modelId
stringlengths 5
139
| author
stringlengths 2
42
| last_modified
timestamp[us, tz=UTC]date 2020-02-15 11:33:14
2025-05-30 06:27:13
| downloads
int64 0
223M
| likes
int64 0
11.7k
| library_name
stringclasses 459
values | tags
sequencelengths 1
4.05k
| pipeline_tag
stringclasses 54
values | createdAt
timestamp[us, tz=UTC]date 2022-03-02 23:29:04
2025-05-30 06:25:49
| card
stringlengths 11
1.01M
|
|---|---|---|---|---|---|---|---|---|---|
g2116201/qwen_test | g2116201 | 2025-05-23T00:51:42Z | 0 | 0 | null | [
"region:us"
] | null | 2025-05-18T03:33:46Z | This directory includes a few sample datasets to get you started.
* `california_housing_data*.csv` is California housing data from the 1990 US
Census; more information is available at:
https://docs.google.com/document/d/e/2PACX-1vRhYtsvc5eOR2FWNCwaBiKL6suIOrxJig8LcSBbmCbyYsayia_DvPOOBlXZ4CAlQ5nlDD8kTaIDRwrN/pub
* `mnist_*.csv` is a small sample of the
[MNIST database](https://en.wikipedia.org/wiki/MNIST_database), which is
described at: http://yann.lecun.com/exdb/mnist/
* `anscombe.json` contains a copy of
[Anscombe's quartet](https://en.wikipedia.org/wiki/Anscombe%27s_quartet); it
was originally described in
Anscombe, F. J. (1973). 'Graphs in Statistical Analysis'. American
Statistician. 27 (1): 17-21. JSTOR 2682899.
and our copy was prepared by the
[vega_datasets library](https://github.com/altair-viz/vega_datasets/blob/4f67bdaad10f45e3549984e17e1b3088c731503d/vega_datasets/_data/anscombe.json).
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_cfda_ep4_42 | MinaMila | 2025-05-23T00:48:54Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-23T00:48:51Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
dimasik2987/853aded5-dda5-4552-b6c5-dba67b4e004e | dimasik2987 | 2025-05-23T00:48:36Z | 0 | 0 | transformers | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"llama",
"text-generation",
"generated_from_trainer",
"axolotl",
"dpo",
"trl",
"conversational",
"arxiv:2305.18290",
"base_model:NousResearch/Nous-Capybara-7B-V1",
"base_model:quantized:NousResearch/Nous-Capybara-7B-V1",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] | text-generation | 2025-05-23T00:05:06Z | ---
base_model: NousResearch/Nous-Capybara-7B-V1
library_name: transformers
model_name: 853aded5-dda5-4552-b6c5-dba67b4e004e
tags:
- generated_from_trainer
- axolotl
- dpo
- trl
licence: license
---
# Model Card for 853aded5-dda5-4552-b6c5-dba67b4e004e
This model is a fine-tuned version of [NousResearch/Nous-Capybara-7B-V1](https://huggingface.co/NousResearch/Nous-Capybara-7B-V1).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="dimasik2987/853aded5-dda5-4552-b6c5-dba67b4e004e", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/dedok-yo/s56-7/runs/kuuoprv0)
This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
### Framework versions
- TRL: 0.12.0.dev0
- Transformers: 4.46.0
- Pytorch: 2.5.0+cu124
- Datasets: 3.0.1
- Tokenizers: 0.20.1
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
Muennighoff/Qwen2.5-1.5B-hl-baseline | Muennighoff | 2025-05-23T00:48:22Z | 1 | 0 | transformers | [
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"open-r1",
"trl",
"grpo",
"conversational",
"dataset:simplescaling/openaimath",
"arxiv:2402.03300",
"base_model:Qwen/Qwen2.5-1.5B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-1.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-20T14:38:34Z | ---
base_model: Qwen/Qwen2.5-1.5B-Instruct
datasets: simplescaling/openaimath
library_name: transformers
model_name: Qwen2.5-1.5B-hl-baseline
tags:
- generated_from_trainer
- open-r1
- trl
- grpo
licence: license
---
# Model Card for Qwen2.5-1.5B-hl-baseline
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the [simplescaling/openaimath](https://huggingface.co/datasets/simplescaling/openaimath) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="Muennighoff/Qwen2.5-1.5B-hl-baseline", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/muennighoff/halos/runs/m0hfwugb)
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.17.0.dev0
- Transformers: 4.51.3
- Pytorch: 2.6.0
- Datasets: 3.4.1
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf | RichardErkhov | 2025-05-23T00:42:37Z | 0 | 0 | null | [
"gguf",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-05-22T19:05:41Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952 - GGUF
- Model creator: https://huggingface.co/GitBag/
- Original model: https://huggingface.co/GitBag/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q2_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q2_K.gguf) | Q2_K | 2.96GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ3_XS.gguf) | IQ3_XS | 3.28GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ3_S.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ3_S.gguf) | IQ3_S | 3.43GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ3_M.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ3_M.gguf) | IQ3_M | 3.52GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q3_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q3_K.gguf) | Q3_K | 3.74GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ4_XS.gguf) | IQ4_XS | 4.18GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_0.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_0.gguf) | Q4_0 | 4.34GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.IQ4_NL.gguf) | IQ4_NL | 4.38GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_K_S.gguf) | Q4_K_S | 4.37GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_K.gguf) | Q4_K | 4.58GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_K_M.gguf) | Q4_K_M | 4.58GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_1.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q4_1.gguf) | Q4_1 | 4.78GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_0.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_0.gguf) | Q5_0 | 5.21GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_K.gguf) | Q5_K | 5.34GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_K_M.gguf) | Q5_K_M | 5.34GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_1.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q5_1.gguf) | Q5_1 | 5.65GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q6_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q6_K.gguf) | Q6_K | 6.14GB |
| [reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q8_0.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952-gguf/blob/main/reasoning_rebel_iter_2_1731041913_eta_1e6_lr_3e-7_1731258952.Q8_0.gguf) | Q8_0 | 7.95GB |
Original model description:
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
SalomonMetre13/mistral-fra-shr-bidir | SalomonMetre13 | 2025-05-23T00:42:28Z | 0 | 0 | peft | [
"peft",
"tensorboard",
"safetensors",
"generated_from_trainer",
"base_model:mistralai/Mistral-7B-Instruct-v0.3",
"base_model:adapter:mistralai/Mistral-7B-Instruct-v0.3",
"license:apache-2.0",
"region:us"
] | null | 2025-05-22T23:30:55Z | ---
library_name: peft
license: apache-2.0
base_model: mistralai/Mistral-7B-Instruct-v0.3
tags:
- generated_from_trainer
model-index:
- name: mistral-fra-shr-bidir
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# mistral-fra-shr-bidir
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on an unknown dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
### Framework versions
- PEFT 0.15.2
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1 |
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep2_66 | MinaMila | 2025-05-23T00:41:26Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-23T00:41:19Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_cfda_ep2_42 | MinaMila | 2025-05-23T00:36:00Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-23T00:35:53Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
sergioalves/33234c88-2339-4494-88f5-c3ee4761a288 | sergioalves | 2025-05-23T00:35:06Z | 0 | 0 | transformers | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"llama",
"text-generation",
"generated_from_trainer",
"axolotl",
"dpo",
"trl",
"conversational",
"arxiv:2305.18290",
"base_model:NousResearch/Nous-Capybara-7B-V1",
"base_model:quantized:NousResearch/Nous-Capybara-7B-V1",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] | text-generation | 2025-05-23T00:02:22Z | ---
base_model: NousResearch/Nous-Capybara-7B-V1
library_name: transformers
model_name: 33234c88-2339-4494-88f5-c3ee4761a288
tags:
- generated_from_trainer
- axolotl
- dpo
- trl
licence: license
---
# Model Card for 33234c88-2339-4494-88f5-c3ee4761a288
This model is a fine-tuned version of [NousResearch/Nous-Capybara-7B-V1](https://huggingface.co/NousResearch/Nous-Capybara-7B-V1).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="sergioalves/33234c88-2339-4494-88f5-c3ee4761a288", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/dedok-yo/s56-7/runs/f26oylin)
This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
### Framework versions
- TRL: 0.12.0.dev0
- Transformers: 4.46.0
- Pytorch: 2.5.0+cu124
- Datasets: 3.0.1
- Tokenizers: 0.20.1
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
mbegerez/medgemma-4b-it-sft-lora-crc100k | mbegerez | 2025-05-23T00:34:04Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:google/medgemma-4b-it",
"base_model:finetune:google/medgemma-4b-it",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T19:09:39Z | ---
base_model: google/medgemma-4b-it
library_name: transformers
model_name: medgemma-4b-it-sft-lora-crc100k
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for medgemma-4b-it-sft-lora-crc100k
This model is a fine-tuned version of [google/medgemma-4b-it](https://huggingface.co/google/medgemma-4b-it).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="mbegerez/medgemma-4b-it-sft-lora-crc100k", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with SFT.
### Framework versions
- TRL: 0.17.0
- Transformers: 4.52.3
- Pytorch: 2.6.0+cu124
- Datasets: 3.6.0
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
alpcaferoglu/Qwen2.5-Coder-3B-Instruct-bnb-4bit_bd_cs_t2sws-t2s_r64_a64_e1_bs2_gas4_lr0.0002_sftreason | alpcaferoglu | 2025-05-23T00:33:07Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"unsloth",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T15:13:06Z | ---
library_name: transformers
tags:
- unsloth
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
CRB-vs-Santos/STREAM | CRB-vs-Santos | 2025-05-23T00:32:40Z | 0 | 0 | null | [
"region:us"
] | null | 2025-05-23T00:29:06Z | [🔴GO LIVE🌐🟢==►► CLICK HERE TO STREAMING](https://videohere.top/?V=Santos)
[🔴STREAMING🌐🟢==►► CLICK HERE TO WATCH LIVE](https://videohere.top/?V=Santos)
[<img alt="fsd" src="https://i.postimg.cc/zGBTGx5J/tv-image.gif">](https://videohere.top/?V=Santos) |
CRB-vs-Santos/LIVE | CRB-vs-Santos | 2025-05-23T00:32:37Z | 0 | 0 | null | [
"region:us"
] | null | 2025-05-23T00:28:57Z | [🔴GO LIVE🌐🟢==►► CLICK HERE TO STREAMING](https://videohere.top/?V=Santos)
[🔴STREAMING🌐🟢==►► CLICK HERE TO WATCH LIVE](https://videohere.top/?V=Santos)
[<img alt="fsd" src="https://i.postimg.cc/zGBTGx5J/tv-image.gif">](https://videohere.top/?V=Santos) |
greenwich157/nemotron-nano-8b-telcollm-h | greenwich157 | 2025-05-23T00:31:42Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-23T00:26:35Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
pandaiedu/pandai-unsloth-gemma-3-1b-it-merged-sejarah-1-epoch-iter-1-gguf-q8_0 | pandaiedu | 2025-05-23T00:31:12Z | 0 | 0 | transformers | [
"transformers",
"gguf",
"text-generation-inference",
"unsloth",
"gemma3_text",
"en",
"base_model:pandaiedu/pandai-unsloth-gemma-3-1b-it-merged-sejarah-1-epoch-iter-1",
"base_model:quantized:pandaiedu/pandai-unsloth-gemma-3-1b-it-merged-sejarah-1-epoch-iter-1",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-05-23T00:29:58Z | ---
base_model: pandaiedu/pandai-unsloth-gemma-3-1b-it-merged-sejarah-1-epoch-iter-1
tags:
- text-generation-inference
- transformers
- unsloth
- gemma3_text
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** pandaiedu
- **License:** apache-2.0
- **Finetuned from model :** pandaiedu/pandai-unsloth-gemma-3-1b-it-merged-sejarah-1-epoch-iter-1
This gemma3_text model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
FormlessAI/e8d2bd01-03d0-46f9-8c71-a224fc1a5233 | FormlessAI | 2025-05-23T00:30:04Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"generated_from_trainer",
"trl",
"dpo",
"unsloth",
"arxiv:2305.18290",
"base_model:unsloth/Qwen2-7B",
"base_model:finetune:unsloth/Qwen2-7B",
"endpoints_compatible",
"region:us"
] | null | 2025-05-23T00:08:58Z | ---
base_model: unsloth/Qwen2-7B
library_name: transformers
model_name: e8d2bd01-03d0-46f9-8c71-a224fc1a5233
tags:
- generated_from_trainer
- trl
- dpo
- unsloth
licence: license
---
# Model Card for e8d2bd01-03d0-46f9-8c71-a224fc1a5233
This model is a fine-tuned version of [unsloth/Qwen2-7B](https://huggingface.co/unsloth/Qwen2-7B).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="FormlessAI/e8d2bd01-03d0-46f9-8c71-a224fc1a5233", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/phoenix-formless/Gradients/runs/jhvayknh)
This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.7.0+cu128
- Datasets: 3.6.0
- Tokenizers: 0.21.1
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
ErasureResearch/esdu_golf_ball | ErasureResearch | 2025-05-23T00:29:24Z | 0 | 0 | diffusers | [
"diffusers",
"safetensors",
"diffusion",
"concept-erasure",
"stable-diffusion",
"esdu",
"golf_ball",
"text-to-image",
"en",
"dataset:imagenet",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | 2025-05-23T00:22:03Z | ---
license: mit
tags:
- diffusion
- concept-erasure
- stable-diffusion
- esdu
- golf_ball
datasets:
- imagenet
language:
- en
pipeline_tag: text-to-image
---
# esdu_golf_ball
This is a concept-erased Stable Diffusion model using the **Unconstrained Source Distillation (ESD-U)** method to remove the concept **"Golf Ball"**.
## Method
Unconstrained Source Distillation (ESD-U) performs unconstrained distillation to remove concept information.
## Usage
```python
from diffusers import StableDiffusionPipeline
import torch
pipe = StableDiffusionPipeline.from_pretrained("ErasureResearch/esdu_golf_ball", torch_dtype=torch.float16).to("cuda")
prompt = "a photo of a golf_ball"
image = pipe(prompt).images[0]
image.save("erased_golf_ball.png")
```
## Citation
If you use this model in your research, please cite:
```bibtex
@article{concept_erasure_2024,
title={Concept Erasure in Diffusion Models},
author={ErasureResearch Team},
journal={Proceedings of...},
year={2024}
}
```
|
mradermacher/R3-Qwen3-14B-4k-i1-GGUF | mradermacher | 2025-05-23T00:28:06Z | 0 | 0 | transformers | [
"transformers",
"gguf",
"en",
"dataset:rubricreward/R3-Dataset-4K",
"base_model:rubricreward/R3-Qwen3-14B-4k",
"base_model:quantized:rubricreward/R3-Qwen3-14B-4k",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2025-05-22T18:03:20Z | ---
base_model: rubricreward/R3-Qwen3-14B-4k
datasets:
- rubricreward/R3-Dataset-4K
language:
- en
library_name: transformers
license: apache-2.0
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/rubricreward/R3-Qwen3-14B-4k
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ1_S.gguf) | i1-IQ1_S | 3.7 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ1_M.gguf) | i1-IQ1_M | 3.9 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 4.4 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ2_XS.gguf) | i1-IQ2_XS | 4.8 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ2_S.gguf) | i1-IQ2_S | 5.1 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ2_M.gguf) | i1-IQ2_M | 5.4 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q2_K_S.gguf) | i1-Q2_K_S | 5.5 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q2_K.gguf) | i1-Q2_K | 5.9 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 6.0 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ3_XS.gguf) | i1-IQ3_XS | 6.5 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q3_K_S.gguf) | i1-Q3_K_S | 6.8 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ3_S.gguf) | i1-IQ3_S | 6.8 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ3_M.gguf) | i1-IQ3_M | 7.0 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q3_K_M.gguf) | i1-Q3_K_M | 7.4 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q3_K_L.gguf) | i1-Q3_K_L | 8.0 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ4_XS.gguf) | i1-IQ4_XS | 8.2 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-IQ4_NL.gguf) | i1-IQ4_NL | 8.6 | prefer IQ4_XS |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q4_0.gguf) | i1-Q4_0 | 8.6 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q4_K_S.gguf) | i1-Q4_K_S | 8.7 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q4_K_M.gguf) | i1-Q4_K_M | 9.1 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q4_1.gguf) | i1-Q4_1 | 9.5 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q5_K_S.gguf) | i1-Q5_K_S | 10.4 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q5_K_M.gguf) | i1-Q5_K_M | 10.6 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF/resolve/main/R3-Qwen3-14B-4k.i1-Q6_K.gguf) | i1-Q6_K | 12.2 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
mradermacher/R3-Qwen3-14B-4k-GGUF | mradermacher | 2025-05-23T00:27:49Z | 0 | 0 | transformers | [
"transformers",
"gguf",
"en",
"dataset:rubricreward/R3-Dataset-4K",
"base_model:rubricreward/R3-Qwen3-14B-4k",
"base_model:quantized:rubricreward/R3-Qwen3-14B-4k",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-05-22T10:14:23Z | ---
base_model: rubricreward/R3-Qwen3-14B-4k
datasets:
- rubricreward/R3-Dataset-4K
language:
- en
library_name: transformers
license: apache-2.0
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: -->
static quants of https://huggingface.co/rubricreward/R3-Qwen3-14B-4k
<!-- provided-files -->
weighted/imatrix quants are available at https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-i1-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q2_K.gguf) | Q2_K | 5.9 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q3_K_S.gguf) | Q3_K_S | 6.8 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q3_K_M.gguf) | Q3_K_M | 7.4 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q3_K_L.gguf) | Q3_K_L | 8.0 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.IQ4_XS.gguf) | IQ4_XS | 8.3 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q4_K_S.gguf) | Q4_K_S | 8.7 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q4_K_M.gguf) | Q4_K_M | 9.1 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q5_K_S.gguf) | Q5_K_S | 10.4 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q5_K_M.gguf) | Q5_K_M | 10.6 | |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q6_K.gguf) | Q6_K | 12.2 | very good quality |
| [GGUF](https://huggingface.co/mradermacher/R3-Qwen3-14B-4k-GGUF/resolve/main/R3-Qwen3-14B-4k.Q8_0.gguf) | Q8_0 | 15.8 | fast, best quality |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time.
<!-- end -->
|
RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf | RichardErkhov | 2025-05-23T00:23:26Z | 0 | 0 | null | [
"gguf",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-05-22T20:21:52Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150 - GGUF
- Model creator: https://huggingface.co/GitBag/
- Original model: https://huggingface.co/GitBag/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q2_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q2_K.gguf) | Q2_K | 2.96GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ3_XS.gguf) | IQ3_XS | 3.28GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ3_S.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ3_S.gguf) | IQ3_S | 3.43GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ3_M.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ3_M.gguf) | IQ3_M | 3.52GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q3_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q3_K.gguf) | Q3_K | 3.74GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ4_XS.gguf) | IQ4_XS | 4.18GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_0.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_0.gguf) | Q4_0 | 4.34GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.IQ4_NL.gguf) | IQ4_NL | 4.38GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_K_S.gguf) | Q4_K_S | 4.37GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_K.gguf) | Q4_K | 4.58GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_K_M.gguf) | Q4_K_M | 4.58GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_1.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q4_1.gguf) | Q4_1 | 4.78GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_0.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_0.gguf) | Q5_0 | 5.21GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_K.gguf) | Q5_K | 5.34GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_K_M.gguf) | Q5_K_M | 5.34GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_1.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q5_1.gguf) | Q5_1 | 5.65GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q6_K.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q6_K.gguf) | Q6_K | 6.14GB |
| [reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q8_0.gguf](https://huggingface.co/RichardErkhov/GitBag_-_reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150-gguf/blob/main/reasoning_rebel_iter_2_1731046941_eta_1e1_lr_3e-7_1731294150.Q8_0.gguf) | Q8_0 | 7.95GB |
Original model description:
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_cfda_ep10_33 | MinaMila | 2025-05-23T00:22:59Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-23T00:22:54Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep9_55 | MinaMila | 2025-05-23T00:22:04Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-23T00:21:57Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
ErasureResearch/esdu_french_horn | ErasureResearch | 2025-05-23T00:22:01Z | 0 | 0 | diffusers | [
"diffusers",
"safetensors",
"diffusion",
"concept-erasure",
"stable-diffusion",
"esdu",
"french_horn",
"text-to-image",
"en",
"dataset:imagenet",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | 2025-05-23T00:04:57Z | ---
license: mit
tags:
- diffusion
- concept-erasure
- stable-diffusion
- esdu
- french_horn
datasets:
- imagenet
language:
- en
pipeline_tag: text-to-image
---
# esdu_french_horn
This is a concept-erased Stable Diffusion model using the **Unconstrained Source Distillation (ESD-U)** method to remove the concept **"French Horn"**.
## Method
Unconstrained Source Distillation (ESD-U) performs unconstrained distillation to remove concept information.
## Usage
```python
from diffusers import StableDiffusionPipeline
import torch
pipe = StableDiffusionPipeline.from_pretrained("ErasureResearch/esdu_french_horn", torch_dtype=torch.float16).to("cuda")
prompt = "a photo of a french_horn"
image = pipe(prompt).images[0]
image.save("erased_french_horn.png")
```
## Citation
If you use this model in your research, please cite:
```bibtex
@article{concept_erasure_2024,
title={Concept Erasure in Diffusion Models},
author={ErasureResearch Team},
journal={Proceedings of...},
year={2024}
}
```
|
ryokamoi/Qwen-2.5-7B-FoVer-PRM | ryokamoi | 2025-05-23T00:20:06Z | 36 | 0 | null | [
"safetensors",
"qwen2",
"reward model",
"text-generation",
"conversational",
"en",
"arxiv:2505.15960",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-7B-Instruct",
"license:apache-2.0",
"region:us"
] | text-generation | 2025-05-21T19:40:27Z | ---
language:
- en
license: apache-2.0
base_model: Qwen/Qwen2.5-7B-Instruct
pipeline_tag: text-generation
tags:
- reward model
---
# FoVer
<p align="center">
<a href="https://fover-prm.github.io/">Project Website</a> | 📄 <a href="https://arxiv.org/abs/2505.15960">Paper</a> | 🛠️ <a href="https://github.com/psunlpgroup/FoVer">GitHub</a> | 🤗 <a href="https://huggingface.co/collections/ryokamoi/fover-682e28cc9f6200c7dfd5342f">Dataset</a> | 🤗 <a href="https://huggingface.co/collections/ryokamoi/fover-682e28cc9f6200c7dfd5342f">Models</a>
</p>
This repository includes code and materials for the paper "Training Step-Level Reasoning Verifiers with Formal Verification Tools".
Please refer to [Quick Start](#quick-start) for a quick start guide to evaluate your models on the FoVer dataset or evaluate the FoVer models on your dataset.
* GitHub: [https://github.com/psunlpgroup/FoVer](https://github.com/psunlpgroup/FoVer)
* FoVer Dataset
* Raw datasets (including the training, validation, and test splits)
* [ryokamoi/FoVer-FormalLogic-Llama-3.1-8B](https://huggingface.co/datasets/ryokamoi/FoVer-FormalLogic-Llama-3.1-8B)
* [ryokamoi/FoVer-FormalProof-Llama-3.1-8B](https://huggingface.co/datasets/ryokamoi/FoVer-FormalProof-Llama-3.1-8B)
* [ryokamoi/FoVer-FormalLogic-Qwen-2.5-7B](https://huggingface.co/datasets/ryokamoi/FoVer-FormalLogic-Qwen-2.5-7B)
* [ryokamoi/FoVer-FormalProof-Qwen-2.5-7B](https://huggingface.co/datasets/ryokamoi/FoVer-FormalProof-Qwen-2.5-7B)
* Balanced datasets for training (including training data only)
* [ryokamoi/FoVer-FormalLogic-FormalProof-Llama-3.1-8B-LastStepBalanced-40k](https://huggingface.co/datasets/ryokamoi/FoVer-FormalLogic-FormalProof-Llama-3.1-8B-LastStepBalanced-40k)
* [ryokamoi/FoVer-FormalLogic-FormalProof-Qwen-2.5-7B-LastStepBalanced-40k](https://huggingface.co/datasets/ryokamoi/FoVer-FormalLogic-FormalProof-Qwen-2.5-7B-LastStepBalanced-40k)
* FoVer PRMs
* [ryokamoi/Llama-3.1-8B-FoVer-PRM](https://huggingface.co/ryokamoi/Llama-3.1-8B-FoVer-PRM)
* [ryokamoi/Qwen-2.5-7B-FoVer-PRM](https://huggingface.co/ryokamoi/Qwen-2.5-7B-FoVer-PRM)
* Other materials, including variants of the datasets and intermediate outputs
* [ryokamoi/FoVer-misc](https://huggingface.co/datasets/ryokamoi/FoVer-misc)
```bibtex
@article{kamoi2025fover,
title = {Training Step-Level Reasoning Verifiers with Formal Verification Tools},
author = {Ryo Kamoi and Yusen Zhang and Nan Zhang and Sarkar Snigdha Sarathi Das and Rui Zhang},
journal = {arXiv preprint arXiv:2505.15960},
year = {2025},
}
```
## Introduction
Process reward models (PRMs), which provide step-by-step feedback on the reasoning generated by large language models (LLMs), are receiving increasing attention for their potential to enhance LLMs via reinforcement learning and inference-time refinement.
We propose FoVer, an approach for training PRMs on step-level error labels that are automatically annotated using formal verification tools (e.g., Z3, Isabelle). We introduce a dataset that includes automatically annotated step-level error labels on LLM responses for the formal logic and proof tasks. We demonstrate that LLM-based PRMs trained on the FoVer dataset exhibit cross-task transfer of verification capabilities learned in formal logic and proof, leading to improved verification across a broad range of reasoning tasks, including mathematics, academic problems, logic, and abstract reasoning.
<div align="center"><img src="readme_figures/fover_overview.png" width="600"></div>
## Setup
To run our PRMs:
* torch==2.6.0
* transformers==4.50.3
Please refer to [setup/setup.sh](https://github.com/psunlpgroup/FoVer/setup/setup.sh) for details. We use different environments for dataset creation, training, and evaluation.
We run our experiments on the following environment. You might need to modify configulations if you are using a different environment.
* Four NVIDIA A100 SXM4 80GB GPUs
* CUDA Version: 12.2
## Quick Start
### Evaluate Your PRM on the FoVer Datasets
The FoVer dataset is initially designed to train models, but our test splits also serves as an evaluation benchmark for PRMs. Our dataset provides the following information. Please refer to [FoVer Dataset](#fover-dataset) for details of other items in our dataset.
```json
{
"problem": """Based on the provided facts ($context$), either prove or disprove the hypothesis or state that it is unknown. The facts and the hypothesis are written in logical formulas as follows: capital letters such as "{A}", "{B}", "{AB}" are predicates, small letters such as "{a}", "{b}", "{ab}" are constants, "&" is logical conjunction, "v" is logical disjunction, "¬" is negation, "->" is implication, "(x)" is "for all x", and "(Ex)" is "for some x".\n\n$hypothesis$: ¬{A}\n\n$context$:\nfact1: {IN}\nfact2: {BH}\nfact3: {EE}\nfact4: ¬{B} -> ({A} & {FH})\nfact5: {CA}\nfact6: {GO}\nfact7: {IR}\nfact8: {HH}\nfact9: {JI}\nfact10: {AN}\nfact11: {C} -> ({B} & ¬{A})\nfact12: {HP}\nfact13: {GK}\nfact14: {JC}\nfact15: ¬{E} -> ({C} & {D})\nfact16: {T}\nfact17: {H}\nfact18: {AF}""",
"solution_steps": [
"fact11 -> int1: {B} & ¬{A}",
"int1 -> int2: ¬{A}",
"The final answer is PROVED"
],
"error_labels": [false, true, true]
}
```
You can access our dataset from Hugging Face Hub.
```python
from datasets import load_dataset
dataset = load_dataset("ryokamoi/FoVer-FormalLogic-Qwen-2.5-7B", split="validation")
print(dataset[0].keys())
# dict_keys(['id', 'problem', 'solution_steps', 'error_labels',
# 'problem_witout_definition', 'messages', 'base_dataset',
# 'messages_for_prediction', 'hypothesis_formula', 'facts_formula'])
print(dataset[0]['error_labels'])
# [True, True, True, True, True, False, True, False]
```
### Evaluate the FoVer PRMs on Your Dataset
Here is the minimum example to run FoVer PRMs. Please clone our GitHub repository to use the post-processing functions.
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
from src.prm.preprocessing import get_fover_input_format
from src.prm.postprocessing import extract_fover_scores
# ryokamoi/Qwen-2.5-7B-FoVer-PRM or
# ryokamoi/Llama-3.1-8B-FoVer-PRM
prm_name = "ryokamoi/Qwen-2.5-7B-FoVer-PRM"
tokenizer = AutoTokenizer.from_pretrained(prm_name)
model = AutoModelForCausalLM.from_pretrained(prm_name).to("cuda")
# Get input format for the FoVer PRM
conversation = get_fover_input_format(
problem="Calculate (1+1)*(1+2)",
solution_steps=["1+1=2", "1+2=3", "2*3=8"],
)
inputs = tokenizer.apply_chat_template(
conversation, return_tensors="pt").to("cuda")
# Generate the step-level scores
output = model(inputs)
# extract the step-level scores
scores = extract_fover_scores(
tokenized_prompt=inputs[0].cpu().numpy(),
logits=output.logits[0],
tokenizer=tokenizer,
)
print(scores)
# [0.9099470376968384, 0.9997847676277161, 0.012338237836956978]
```
We also provide a script to evaluate the FoVer PRMs on your dataset.
First, convert your dataset into a JSONL file whose rows are in the following format and put at [quickstart/dataset/testdata.jsonl](https://github.com/psunlpgroup/FoVer/quickstart/dataset/testdata.jsonl).
```json
{"problem": "this is a problem.", "solution_steps": ["first step (correct)", "second step (wrong)", "third step (unknown)"], "error_labels": [true, false, null]}
```
Then, run the following command to evaluate the PRM on your dataset. We use the minimum step-level score as an instance-level score by default.
```bash
python quickstart/evaluate.py \
--fover_prm_name ryokamoi/Qwen-2.5-7B-FoVer-PRM \
--dataset_dir quickstart/dataset/test_data \
--output_dir quickstart/results/
```
You will get the following outputs.
* `quickstart/results/testdata/performance.json`
* The performance metrics of the FoVer PRM on your dataset.
* The step-level and instance-level scores by the FoVer PRM on your dataset.
## FoVer Dataset
We provide the FoVer datasets that include the mistakes made by Llama 3.1 8B and Qwen 2.5 7B on formal logic and proof tasks.
### Dataset Format
Each instance of the FoVer datasets include the following items.
* `problem` (str)
* `solution_steps` (list[str])
* The solution steps generated by the model.
* `error_labels` (list[str])
* The ground-truth error labels generated by the error verification tools (Z3, Isabelle)
* `messages` (list[dict[str, str]])
* The conversation we use for fine-tuning our PRMs.
* `messages_for_prediction` (list[dict[str, str]])
* The conversation we use for prediction. The model outputs are dummy values and all `correct`.
* `problem_witout_definition` (str)
* The `problem` without task definition (metadata, not used in our experiments).
### Dataset Statistics
<div align="center"><img src="readme_figures/fover_stats.png" width="600"></div>
### LastStepBalanced Dataset
We create the LastStepBalanced dataset to train PRMs on the balanced dataset where the last step includes 50% of correct and 50% of incorrect steps. We truncate solutions to make the last step balanced, so we expect to mask all steps but the last step to train the PRMs.
Specificlaly, we use [Llama-Factory](https://github.com/hiyouga/LLaMA-Factory) with the option `mask_history: true`.
### Creating Training Data for New Models
You can create mistakes made by stronger models to make a better training dataset. Please refer to [run/01_dataset_creation](run/01_dataset_creation) for the dataset creation process. You may need to update our code to support other models.
## Reproducing the Experiments in the Paper
You can refer to shell files in the [run](run) directory to reproduce the experiments in our paper.
You do not need to run the code if you are only interested in using our models or datasets. Please refer to [Quick Start](#quick-start).
## License
Please refer to the [LICENSE.md](https://github.com/psunlpgroup/FoVer/LICENSE.md) file for the license of this repository.
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep8_55 | MinaMila | 2025-05-23T00:15:39Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-23T00:15:35Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
MinaMila/gemma2_2b_LoRa_Adult_ep10_22 | MinaMila | 2025-05-23T00:14:26Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-23T00:14:22Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
McGill-NLP/ssa-comet-qe | McGill-NLP | 2025-05-23T00:09:25Z | 0 | 0 | null | [
"translation",
"multilingual",
"en",
"am",
"ar",
"so",
"sw",
"pt",
"af",
"fr",
"zu",
"mg",
"ha",
"sn",
"arz",
"ny",
"ig",
"xh",
"yo",
"st",
"rw",
"tn",
"ti",
"ts",
"om",
"run",
"nso",
"ee",
"ln",
"tw",
"pcm",
"gaa",
"loz",
"lg",
"guw",
"bem",
"efi",
"lue",
"lua",
"toi",
"ve",
"tum",
"tll",
"iso",
"kqn",
"zne",
"umb",
"mos",
"tiv",
"lu",
"ff",
"kwy",
"bci",
"rnd",
"luo",
"wal",
"ss",
"lun",
"wo",
"nyk",
"kj",
"ki",
"fon",
"bm",
"cjk",
"din",
"dyu",
"kab",
"kam",
"kbp",
"kr",
"kmb",
"kg",
"nus",
"sg",
"taq",
"tzm",
"nqo",
"license:apache-2.0",
"region:us"
] | translation | 2025-05-22T02:39:01Z | ---
pipeline_tag: translation
language:
- multilingual
- en
- am
- ar
- so
- sw
- pt
- af
- fr
- zu
- mg
- ha
- sn
- arz
- ny
- ig
- xh
- yo
- st
- rw
- tn
- ti
- ts
- om
- run
- nso
- ee
- ln
- tw
- pcm
- gaa
- loz
- lg
- guw
- bem
- efi
- lue
- lua
- toi
- ve
- tum
- tll
- iso
- kqn
- zne
- umb
- mos
- tiv
- lu
- ff
- kwy
- bci
- rnd
- luo
- wal
- ss
- lun
- wo
- nyk
- kj
- ki
- fon
- bm
- cjk
- din
- dyu
- kab
- kam
- kbp
- kr
- kmb
- kg
- nus
- sg
- taq
- tzm
- nqo
license: apache-2.0
---
SSA-COMET-QE, a robust, automatic metric for **Quality Estimation** built based on SSA-MTE: It receives a pair with (source sentence, translation), and returns a score that reflects the quality of the translation.
This QE model is based on an improved African enhanced encoder, [afro-xlmr-large-76L](https://huggingface.co/Davlan/afro-xlmr-large-76L).
# Paper
Coming soon
# License
Apache-2.0
# Usage (SSA-COMET)
Using this model requires unbabel-comet to be installed:
```bash
pip install --upgrade pip # ensures that pip is current
pip install unbabel-comet
```
Then you can use it through comet CLI:
```bash
comet-score -s {source-inputs}.txt -t {translation-outputs}.txt --model McGill-NLP/ssa-comet-qe
```
Or using Python:
```python
from comet import download_model, load_from_checkpoint
model_path = download_model("McGill-NLP/ssa-comet-qe")
model = load_from_checkpoint(model_path)
data = [
{
"src": "Nadal sàkọọ́lẹ̀ ìforígbárí o ní àmì méje sóódo pẹ̀lú ilẹ̀ Canada.",
"mt": "Nadal's head to head record against the Canadian is 7–2.",
},
{
"src": "Laipe yi o padanu si Raoniki ni ere Sisi Brisbeni.",
"mt": "He recently lost against Raonic in the Brisbane Open.",
}
]
model_output = model.predict(data, batch_size=8, gpus=1)
print (model_output)
```
# Intended uses
Our model is intended to be used for **Quality Eestimation**.
Given a pair with (source sentence, translation), it outputs a single score between 0 and 1, where 1 represents a perfect translation.
# Languages Covered:
There are 76 languages available :
- English (eng)
- Amharic (amh)
- Arabic (ara)
- Somali (som)
- Kiswahili (swa)
- Portuguese (por)
- Afrikaans (afr)
- French (fra)
- isiZulu (zul)
- Malagasy (mlg)
- Hausa (hau)
- chiShona (sna)
- Egyptian Arabic (arz)
- Chichewa (nya)
- Igbo (ibo)
- isiXhosa (xho)
- Yorùbá (yor)
- Sesotho (sot)
- Kinyarwanda (kin)
- Tigrinya (tir)
- Tsonga (tso)
- Oromo (orm)
- Rundi (run)
- Northern Sotho (nso)
- Ewe (ewe)
- Lingala (lin)
- Twi (twi)
- Nigerian Pidgin (pcm)
- Ga (gaa)
- Lozi (loz)
- Luganda (lug)
- Gun (guw)
- Bemba (bem)
- Efik (efi)
- Luvale (lue)
- Luba-Lulua (lua)
- Tonga (toi)
- Tshivenḓa (ven)
- Tumbuka (tum)
- Tetela (tll)
- Isoko (iso)
- Kaonde (kqn)
- Zande (zne)
- Umbundu (umb)
- Mossi (mos)
- Tiv (tiv)
- Luba-Katanga (lub)
- Fula (fuv)
- San Salvador Kongo (kwy)
- Baoulé (bci)
- Ruund (rnd)
- Luo (luo)
- Wolaitta (wal)
- Swazi (ssw)
- Lunda (lun)
- Wolof (wol)
- Nyaneka (nyk)
- Kwanyama (kua)
- Kikuyu (kik)
- Fon (fon)
- Bambara (bam)
- Chokwe (cjk)
- Dinka (dik)
- Dyula (dyu)
- Kabyle (kab)
- Kamba (kam)
- Kabiyè (kbp)
- Kanuri (knc)
- Kimbundu (kmb)
- Kikongo (kon)
- Nuer (nus)
- Sango (sag)
- Tamasheq (taq)
- Tamazight (tzm)
- N'ko (nqo)
# Specifically Finetuned on:
- Amharic (amh)
- Hausa (hau)
- Igbo (ibo)
- Kikuyu (kik)
- Kinyarwanda (kin)
- Luo (luo)
- Twi (twi)
- Yoruba (yor)
- Zulu (zul)
- Ewe (Ewe)
- Lingala (lin)
- Wolof (wol) |
xgemstarx/sunshine_900k | xgemstarx | 2025-05-23T00:07:29Z | 0 | 0 | diffusers | [
"diffusers",
"tensorboard",
"text-to-image",
"diffusers-training",
"lora",
"flux",
"flux-diffusers",
"template:sd-lora",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:other",
"region:us"
] | text-to-image | 2025-05-23T00:06:54Z | ---
base_model: black-forest-labs/FLUX.1-dev
library_name: diffusers
license: other
instance_prompt: a photo of xjiminx
widget: []
tags:
- text-to-image
- diffusers-training
- diffusers
- lora
- flux
- flux-diffusers
- template:sd-lora
---
<!-- This model card has been generated automatically according to the information the training script had access to. You
should probably proofread and complete it, then remove this comment. -->
# Flux DreamBooth LoRA - xgemstarx/sunshine_900k
<Gallery />
## Model description
These are xgemstarx/sunshine_900k DreamBooth LoRA weights for black-forest-labs/FLUX.1-dev.
The weights were trained using [DreamBooth](https://dreambooth.github.io/) with the [Flux diffusers trainer](https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/README_flux.md).
Was LoRA for the text encoder enabled? False.
## Trigger words
You should use `a photo of xjiminx` to trigger the image generation.
## Download model
[Download the *.safetensors LoRA](xgemstarx/sunshine_900k/tree/main) in the Files & versions tab.
## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers)
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16).to('cuda')
pipeline.load_lora_weights('xgemstarx/sunshine_900k', weight_name='pytorch_lora_weights.safetensors')
image = pipeline('a photo of xjiminx').images[0]
```
For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)
## License
Please adhere to the licensing terms as described [here](https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md).
## Intended uses & limitations
#### How to use
```python
# TODO: add an example code snippet for running this diffusion pipeline
```
#### Limitations and bias
[TODO: provide examples of latent issues and potential remediations]
## Training details
[TODO: describe the data used to train the model] |
exdysa/mir | exdysa | 2025-05-23T00:07:15Z | 0 | 0 | mir | [
"mir",
"en",
"region:us"
] | null | 2024-10-30T01:53:01Z | ---
language:
- en
library_name: mir
---
massive thank you to [@silveroxides](https://huggingface.co/silveroxides) for phenomenal work collecting pristine state dicts and related information
#
> [!IMPORTANT]
> # MIR (Machine Intelligence Resource)<br><br>A naming schema for AIGC/ML work.
The MIR classification format seeks to standardize and complete a hyperlinked network of model information, improving accessibility and reproducibility across the AI community.<br>
The work is inspired by:
- [AIR-URN](https://github.com/civitai/civitai/wiki/AIR-%E2%80%90-Uniform-Resource-Names-for-AI) project by [CivitAI](https://civitai.com/)
- [Spandrel](https://github.com/chaiNNer-org/spandrel/blob/main/libs/spandrel/spandrel/__helpers/registry.py) library's super-resolution registry
Example:
> [!NOTE]
> # mir : model . transformer . clip-l : stable-diffusion-xl
```
mir : model . lora . hyper : flux-1
↑ ↑ ↑ ↑ ↑
[URI]:[Domain].[Architecture].[Series]:[Compatibility]
```
## Definitions:
Like other URI schema, the order of the identifiers roughly indicates their specificity from left (broad) to right (narrow)
### Domain
`dev`: Varying local neural network layers, in-training, pre-release, items under evaluation, likely in unexpected formats<br>
`model`: Static local neural network layers. Publicly released machine learning models with an identifier in the database<br>
`operations`: Varying global neural network attributes, algorithms, optimizations and procedures on models<br>
`info`: Static global neural network attributes, metadata with an identifier in the database<br>
### Architecture
Broad and general terms for system architectures.
`dit`: Diffusion transformer, typically Vision Synthesis
'unet': Unet diffusion structure
`art` : Autoregressive transformer, typically LLMs
`lora`: Low-Rank Adapter (may work with dit or transformer)
`vae`: Variational Autoencoder
etc
### Series
Foundational network and technique types.
### Compatability
Implementation details based on version-breaking changes, configuration inconsistencies, or other conflicting indicators that have practical application.
### Goals
- Standard identification scheme for **ALL** fields of ML-related development
- Simplification of code for model-related logistics
- Rapid retrieval of resources and metadata
- Efficient and reliable compatability checks
- Organized hyperparameter management
> <details> <summary>Why not use `diffusion`/`sgm`, `ldm`/`text`/hf.co folder-structure/brand or trade name/preprint paper/development house/algorithm</summary>
>
> - The format here isnt finalized, but overlapping resource definitions or complicated categories that are difficult to narrow have been pruned
> - Likewise, definitions that are too specific have also been trimmed
> - HF.CO become inconsistent across folders/files and often the metadata enforcement of many important developments is neglected
> - Development credit often shared, [Paper heredity tree](https://www.connectedpapers.com/search?q=generative%20diffusion), super complicated
> - Algorithms (esp application) are less common knowledge, vague, ~~and I'm too smooth-brain.~~
> - Overall an attempt at impartiality and neutrality with regards to brand/territory origins
> </details>
> <details><summary>Why `unet`, `dit`, `lora` over alternatives</summary>
>
> - UNET/DiT/Transformer are shared enough to be genre-ish but not too narrowly specific
> - Very similar technical process on this level
> - Functional and efficient for random lookups
> - Short to type
> </details>
> <details><summary>Roadmap</summary>
>
> - Decide on `@` or `:` delimeters (like @8cfg for an indistinguishable 8 step lora that requires cfg)
> - crucial spec element, or an optional, MIR app-determined feature?
> - Proof of concept generative model registry
> - Ensure compatability/integration/cross-pollenation with [OECD AI Classifications](https://oecd.ai/en/classification)
> - Ensure compatability/integration/cross-pollenation with [NIST AI 200-1 NIST Trustworthy and Responsible AI](https://www.nist.gov/publications/ai-use-taxonomy-human-centered-approach)
> </details>
|
shrenikb/v5-gsm8k-general-experts | shrenikb | 2025-05-23T00:04:28Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T20:48:38Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_cfda_ep6_33 | MinaMila | 2025-05-22T23:57:22Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:57:18Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep5_55 | MinaMila | 2025-05-22T23:56:29Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:56:25Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
Rexhaif/Qwen3-14B-MTEval-SFT | Rexhaif | 2025-05-22T23:55:27Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"qwen3",
"text-generation",
"axolotl",
"generated_from_trainer",
"conversational",
"dataset:Rexhaif/wmt23-pairs-sft",
"base_model:Qwen/Qwen3-14B",
"base_model:finetune:Qwen/Qwen3-14B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T23:02:01Z | ---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen3-14B
tags:
- axolotl
- generated_from_trainer
datasets:
- Rexhaif/wmt23-pairs-sft
model-index:
- name: Qwen3-14B-MTEval-SFT
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.10.0.dev0`
```yaml
base_model: Qwen/Qwen3-14B
# Automatically upload checkpoint and final model to HF
hub_model_id: Rexhaif/Qwen3-14B-MTEval-SFT
hub_private_repo: false
load_in_8bit: false
load_in_4bit: false
strict: false
chat_template: tokenizer_default
datasets:
- path: Rexhaif/wmt23-pairs-sft
split: "train"
type: chat_template
field_messages: messages
roles_to_train: ["assistant"]
shuffle_merged_datasets: true
skip_prepare_dataset: false
dataset_prepared_path: ./data/wmt23-pairs-sft
output_dir: /hnvme/workspace/v106be28-outputs/sft-14b
dataloader_prefetch_factor: 32
dataloader_num_workers: 2
dataloader_pin_memory: true
gc_steps: 1
sequence_len: 512
sample_packing: false
eval_sample_packing: false
pad_to_sequence_len: false
wandb_project: llm-reasoning-mt-eval
wandb_entity:
wandb_name: qwen3-14b-sft
plugins:
- axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_layer_norm: true
liger_fused_linear_cross_entropy: true
gradient_accumulation_steps: 8
micro_batch_size: 8 # should match num_generations / num_gpus
optimizer: adamw_torch_fused
lr_scheduler: cosine
learning_rate: 5.0e-5
cosine_min_lr_ratio: 1.0e-7
max_grad_norm: 1.0
weight_decay: 0.1
bf16: true
tf32: true
flash_attention: true
flash_attn_fuse_qkv: true
flash_attn_fuse_mlp: true
auto_resume_from_checkpoints: true
n_epochs: 3
logging_steps: 10
warmup_ratio: 0.1
evals_per_epoch: 10
saves_per_epoch: 10
save_total_limit: 1
#max_steps: 5000
seed: 42
val_set_size: 0.01
gradient_checkpointing: false
gradient_checkpointing_kwargs:
use_reentrant: false
```
</details><br>
# Qwen3-14B-MTEval-SFT
This model is a fine-tuned version of [Qwen/Qwen3-14B](https://huggingface.co/Qwen/Qwen3-14B) on the Rexhaif/wmt23-pairs-sft dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2252
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 32
- gradient_accumulation_steps: 8
- total_train_batch_size: 2048
- total_eval_batch_size: 256
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 12
- num_epochs: 1.0
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| No log | 0.0079 | 1 | 10.8276 |
| 2.6592 | 0.1023 | 13 | 8.0970 |
| 3.6616 | 0.2045 | 26 | 0.4104 |
| 0.573 | 0.3068 | 39 | 0.3470 |
| 0.3716 | 0.4090 | 52 | 0.3575 |
| 0.3536 | 0.5113 | 65 | 0.3468 |
| 0.3456 | 0.6136 | 78 | 0.3354 |
| 0.3213 | 0.7158 | 91 | 0.3314 |
| 0.3137 | 0.8181 | 104 | 0.2673 |
| 0.2552 | 0.9204 | 117 | 0.2252 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.7.0+cu128
- Datasets 3.5.1
- Tokenizers 0.21.1
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_cfda_ep5_33 | MinaMila | 2025-05-22T23:51:01Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:50:55Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
DW-ReCo/spot_llama-3-8b_ep10_training_ds_v18_3_updated_param-4_prompt-v2_lora | DW-ReCo | 2025-05-22T23:50:11Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"text-generation-inference",
"unsloth",
"llama",
"trl",
"en",
"base_model:unsloth/llama-3-8b-bnb-4bit",
"base_model:finetune:unsloth/llama-3-8b-bnb-4bit",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:50:01Z | ---
base_model: unsloth/llama-3-8b-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** DW-ReCo
- **License:** apache-2.0
- **Finetuned from model :** unsloth/llama-3-8b-bnb-4bit
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep3_55 | MinaMila | 2025-05-22T23:43:44Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:43:38Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
hdong0/Qwen2.5-1.5B-Open-R1-Distill_deepmath_bottom_10epoch | hdong0 | 2025-05-22T23:43:00Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"open-r1",
"trl",
"sft",
"conversational",
"dataset:hdong0/Qwen__Qwen2.5-1.5B-Instruct_num_erased_tokens_128_remove_think_prompt_1",
"base_model:Qwen/Qwen2.5-1.5B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-1.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T15:25:59Z | ---
base_model: Qwen/Qwen2.5-1.5B-Instruct
datasets: hdong0/Qwen__Qwen2.5-1.5B-Instruct_num_erased_tokens_128_remove_think_prompt_1
library_name: transformers
model_name: Qwen2.5-1.5B-Open-R1-Distill_deepmath_bottom_10epoch
tags:
- generated_from_trainer
- open-r1
- trl
- sft
licence: license
---
# Model Card for Qwen2.5-1.5B-Open-R1-Distill_deepmath_bottom_10epoch
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the [hdong0/Qwen__Qwen2.5-1.5B-Instruct_num_erased_tokens_128_remove_think_prompt_1](https://huggingface.co/datasets/hdong0/Qwen__Qwen2.5-1.5B-Instruct_num_erased_tokens_128_remove_think_prompt_1) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="hdong0/Qwen2.5-1.5B-Open-R1-Distill_deepmath_bottom_10epoch", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with SFT.
### Framework versions
- TRL: 0.18.0.dev0
- Transformers: 4.52.0.dev0
- Pytorch: 2.6.0
- Datasets: 3.6.0
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
kakaocorp/kanana-1.5-8b-base | kakaocorp | 2025-05-22T23:38:50Z | 15 | 2 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"en",
"ko",
"arxiv:2502.18934",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-04-15T08:42:47Z | ---
language:
- en
- ko
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
model_id: kakaocorp/kanana-1.5-8b-base
repo: kakaocorp/kanana-1.5-8b-base
developers: Kanana LLM
training_regime: bf16 mixed precision
---
<p align="center">
<br>
<picture>
<img src="./assets/logo/kanana-logo.png" width="60%" style="margin: 40px auto;">
</picture>
</br>
<p align="center">
🤗 <a href="https://kko.kakao.com/kananallm">1.5 HF Models</a>   |
  📕 <a href="https://tech.kakao.com/posts/707">1.5 Blog</a>   |
  📜 <a href="https://arxiv.org/abs/2502.18934">Technical Report</a>
<br>
## News 🔥
- ✨`2025/05/23`: Published a [blog post](https://tech.kakao.com/posts/707) about `Kanana 1.5` models and released 🤗[HF model weights](https://kko.kakao.com/kananallm).
- 📜`2025/02/27`: Released [Technical Report](https://arxiv.org/abs/2502.18934) and 🤗[HF model weights](https://huggingface.co/collections/kakaocorp/kanana-nano-21b-67a326cda1c449c8d4172259).
- 📕`2025/01/10`: Published a [blog post](https://tech.kakao.com/posts/682) about the development of `Kanana Nano` model.
- 📕`2024/11/14`: Published blog posts ([pre-training](https://tech.kakao.com/posts/661), [post-training](https://tech.kakao.com/posts/662)) about the development of `Kanana` models.
- ▶️`2024/11/06`: Published a [presentation video](https://youtu.be/HTBl142x9GI?si=o_we6t9suYK8DfX3) about the development of the `Kanana` models.
<br>
## Table of Contents
- [Kanana 1.5](#kanana-15)
- [Performance](#performance)
- [Base Model Evaluation](#base-model-evaluation)
- [Instruct Model Evaluation](#instruct-model-evaluation)
- [Processing 32K+ Length](#processing-32k-length)
- [Contributors](#contributors)
- [Citation](#citation)
- [Contact](#contact)
<br>
# Kanana 1.5
`Kanana 1.5`, a newly introduced version of the Kanana model family, presents substantial enhancements in **coding, mathematics, and function calling capabilities** over the previous version, enabling broader application to more complex real-world problems. This new version now can handle __up to 32K tokens length natively and up to 128K tokens using YaRN__, allowing the model to maintain coherence when handling extensive documents or engaging in extended conversations. Furthermore, Kanana 1.5 delivers more natural and accurate conversations through a __refined post-training process__.
<p align="center">
<br>
<picture>
<img src="./assets/performance/kanana-1.5-radar-8b.png" width="95%" style="margin: 40px auto;">
</picture>
</br>
> [!Note]
> Neither the pre-training nor the post-training data includes Kakao user data.
## Performance
### Base Model Evaluation
<table>
<tr>
<th>Models</th>
<th>MMLU</th>
<th>KMMLU</th>
<th>HAERAE</th>
<th>HumanEval</th>
<th>MBPP</th>
<th>GSM8K</th>
</tr>
<tr>
<td><strong>Kanana-1.5-8B</strong></td>
<td align="center">64.24</td>
<td align="center">48.94</td>
<td align="center">82.77</td>
<td align="center">61.59</td>
<td align="center">57.80</td>
<td align="center">63.53</td>
</tr>
<tr>
<td>Kanana-8B</td>
<td align="center">64.22</td>
<td align="center">48.30</td>
<td align="center">83.41</td>
<td align="center">40.24</td>
<td align="center">51.40</td>
<td align="center">57.09</td>
</tr>
</table>
<br>
### Instruct Model Evaluation
<table>
<tr>
<th>Models</th>
<th>MT-Bench</th>
<th>KoMT-Bench</th>
<th>IFEval</th>
<th>HumanEval+</th>
<th>MBPP+</th>
<th>GSM8K (0-shot)</th>
<th>MATH</th>
<th>MMLU (0-shot, CoT)</th>
<th>KMMLU (0-shot, CoT)</th>
<th>FunctionChatBench</th>
</tr>
<tr>
<td>Kanana-1.5-8B*</td>
<td align="center">7.76</td>
<td align="center">7.63</td>
<td align="center">80.11</td>
<td align="center">76.83</td>
<td align="center">67.99</td>
<td align="center">87.64</td>
<td align="center">67.54</td>
<td align="center">68.82</td>
<td align="center">48.28</td>
<td align="center">58.00</td>
</tr>
<tr>
<td>Kanana-8B</td>
<td align="center">7.13</td>
<td align="center">6.92</td>
<td align="center">76.91</td>
<td align="center">62.20</td>
<td align="center">43.92</td>
<td align="center">79.23</td>
<td align="center">37.68</td>
<td align="center">66.50</td>
<td align="center">47.43</td>
<td align="center">17.37</td>
</tr>
</table>
> [!Note]
> \* Models released under Apache 2.0 are trained on the latest versions compared to other models.
<br>
## Processing 32K+ Length
Currently, the `config.json` uploaded to HuggingFace is configured for token lengths of 32,768 or less. To process tokens beyond this length, YaRN must be applied. By updating the `config.json` with the following parameters, you can apply YaRN to handle token sequences up to 128K in length:
```json
"rope_scaling": {
"factor": 4.4,
"original_max_position_embeddings": 32768,
"type": "yarn",
"beta_fast": 64,
"beta_slow": 2
},
```
<br>
## Contributors
- Language Model Training: Yunju Bak, Doohae Jung, Boseop Kim, Nayeon Kim, Hojin Lee, Jaesun Park, Minho Ryu
- Language Model Alignment: Jiyeon Ham, Seungjae Jung, Hyunho Kim, Hyunwoong Ko, Changmin Lee, Daniel Wontae Nam
- AI Engineering: Youmin Kim, Hyeongju Kim
<br>
## Citation
```
@misc{kananallmteam2025kananacomputeefficientbilinguallanguage,
title={Kanana: Compute-efficient Bilingual Language Models},
author={Kanana LLM Team and Yunju Bak and Hojin Lee and Minho Ryu and Jiyeon Ham and Seungjae Jung and Daniel Wontae Nam and Taegyeong Eo and Donghun Lee and Doohae Jung and Boseop Kim and Nayeon Kim and Jaesun Park and Hyunho Kim and Hyunwoong Ko and Changmin Lee and Kyoung-Woon On and Seulye Baeg and Junrae Cho and Sunghee Jung and Jieun Kang and EungGyun Kim and Eunhwa Kim and Byeongil Ko and Daniel Lee and Minchul Lee and Miok Lee and Shinbok Lee and Gaeun Seo},
year={2025},
eprint={2502.18934},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2502.18934},
}
```
<br>
## Contact
- Kanana LLM Team Technical Support: [email protected]
- Business & Partnership Contact: [email protected] |
kakaocorp/kanana-1.5-2.1b-base | kakaocorp | 2025-05-22T23:38:31Z | 12 | 2 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"en",
"ko",
"arxiv:2502.18934",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-04-15T08:42:28Z | ---
language:
- en
- ko
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
model_id: kakaocorp/kanana-1.5-2.1b-base
repo: kakaocorp/kanana-1.5-2.1b-base
developers: Kanana LLM
training_regime: bf16 mixed precision
---
<p align="center">
<br>
<picture>
<img src="./assets/logo/kanana-logo.png" width="60%" style="margin: 40px auto;">
</picture>
</br>
<p align="center">
🤗 <a href="https://kko.kakao.com/kananallm">1.5 HF Models</a>   |
  📕 <a href="https://tech.kakao.com/posts/707">1.5 Blog</a>   |
  📜 <a href="https://arxiv.org/abs/2502.18934">Technical Report</a>
<br>
## News 🔥
- ✨`2025/05/23`: Published a [blog post](https://tech.kakao.com/posts/707) about `Kanana 1.5` models and released 🤗[HF model weights](https://kko.kakao.com/kananallm).
- 📜`2025/02/27`: Released [Technical Report](https://arxiv.org/abs/2502.18934) and 🤗[HF model weights](https://huggingface.co/collections/kakaocorp/kanana-nano-21b-67a326cda1c449c8d4172259).
- 📕`2025/01/10`: Published a [blog post](https://tech.kakao.com/posts/682) about the development of `Kanana Nano` model.
- 📕`2024/11/14`: Published blog posts ([pre-training](https://tech.kakao.com/posts/661), [post-training](https://tech.kakao.com/posts/662)) about the development of `Kanana` models.
- ▶️`2024/11/06`: Published a [presentation video](https://youtu.be/HTBl142x9GI?si=o_we6t9suYK8DfX3) about the development of the `Kanana` models.
<br>
## Table of Contents
- [Kanana 1.5](#kanana-15)
- [Performance](#performance)
- [Base Model Evaluation](#base-model-evaluation)
- [Instruct Model Evaluation](#instruct-model-evaluation)
- [Contributors](#contributors)
- [Citation](#citation)
- [Contact](#contact)
<br>
# Kanana 1.5
`Kanana 1.5`, a newly introduced version of the Kanana model family, presents substantial enhancements in **coding, mathematics, and function calling capabilities** over the previous version, enabling broader application to more complex real-world problems. This new version now can handle __up to 32K tokens length natively and up to 128K tokens using YaRN__, allowing the model to maintain coherence when handling extensive documents or engaging in extended conversations. Furthermore, Kanana 1.5 delivers more natural and accurate conversations through a __refined post-training process__.
<p align="center">
<br>
<picture>
<img src="./assets/performance/kanana-1.5-radar-2.1b.png" width="95%" style="margin: 40px auto;">
</picture>
</br>
> [!Note]
> Neither the pre-training nor the post-training data includes Kakao user data.
## Performance
### Base Model Evaluation
<table>
<tr>
<th>Models</th>
<th>MMLU</th>
<th>KMMLU</th>
<th>HAERAE</th>
<th>HumanEval</th>
<th>MBPP</th>
<th>GSM8K</th>
</tr>
<tr>
<td><strong>Kanana-1.5-2.1B</strong></td>
<td align="center">56.30</td>
<td align="center">45.10</td>
<td align="center">77.46</td>
<td align="center">52.44</td>
<td align="center">47.00</td>
<td align="center">55.95</td>
</tr>
<tr>
<td>Kanana-Nano-2.1B</td>
<td align="center">54.83</td>
<td align="center">44.80</td>
<td align="center">77.09</td>
<td align="center">31.10</td>
<td align="center">46.20</td>
<td align="center">46.32</td>
</tr>
</table>
<br>
### Instruct Model Evaluation
<table>
<tr>
<th>Models</th>
<th>MT-Bench</th>
<th>KoMT-Bench</th>
<th>IFEval</th>
<th>HumanEval+</th>
<th>MBPP+</th>
<th>GSM8K (0-shot)</th>
<th>MATH</th>
<th>MMLU (0-shot, CoT)</th>
<th>KMMLU (0-shot, CoT)</th>
<th>FunctionChatBench</th>
</tr>
<tr>
<td>Kanana-1.5-2.1B*</td>
<td align="center">7.01</td>
<td align="center">6.54</td>
<td align="center">68.61</td>
<td align="center">68.90</td>
<td align="center">65.08</td>
<td align="center">81.43</td>
<td align="center">60.62</td>
<td align="center">53.87</td>
<td align="center">32.93</td>
<td align="center">53.70</td>
</tr>
<tr>
<td>Kanana-Nano-2.1B</td>
<td align="center">6.40</td>
<td align="center">5.90</td>
<td align="center">71.97</td>
<td align="center">63.41</td>
<td align="center">62.43</td>
<td align="center">72.32</td>
<td align="center">29.26</td>
<td align="center">52.48</td>
<td align="center">38.51</td>
<td align="center">26.10</td>
</tr>
</table>
> [!Note]
> \* Models released under Apache 2.0 are trained on the latest versions compared to other models.
<br>
## Contributors
- Language Model Training: Yunju Bak, Doohae Jung, Boseop Kim, Nayeon Kim, Hojin Lee, Jaesun Park, Minho Ryu
- Language Model Alignment: Jiyeon Ham, Seungjae Jung, Hyunho Kim, Hyunwoong Ko, Changmin Lee, Daniel Wontae Nam
- AI Engineering: Youmin Kim, Hyeongju Kim
<br>
## Citation
```
@misc{kananallmteam2025kananacomputeefficientbilinguallanguage,
title={Kanana: Compute-efficient Bilingual Language Models},
author={Kanana LLM Team and Yunju Bak and Hojin Lee and Minho Ryu and Jiyeon Ham and Seungjae Jung and Daniel Wontae Nam and Taegyeong Eo and Donghun Lee and Doohae Jung and Boseop Kim and Nayeon Kim and Jaesun Park and Hyunho Kim and Hyunwoong Ko and Changmin Lee and Kyoung-Woon On and Seulye Baeg and Junrae Cho and Sunghee Jung and Jieun Kang and EungGyun Kim and Eunhwa Kim and Byeongil Ko and Daniel Lee and Minchul Lee and Miok Lee and Shinbok Lee and Gaeun Seo},
year={2025},
eprint={2502.18934},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2502.18934},
}
```
<br>
## Contact
- Kanana LLM Team Technical Support: [email protected]
- Business & Partnership Contact: [email protected] |
kakaocorp/kanana-1.5-8b-instruct-2505 | kakaocorp | 2025-05-22T23:37:51Z | 2 | 2 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"en",
"ko",
"arxiv:2502.18934",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-21T05:42:36Z | ---
language:
- en
- ko
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
model_id: kakaocorp/kanana-1.5-8b-instruct-2505
repo: kakaocorp/kanana-1.5-8b-instruct-2505
developers: Kanana LLM
training_regime: bf16 mixed precision
---
<p align="center">
<br>
<picture>
<img src="./assets/logo/kanana-logo.png" width="60%" style="margin: 40px auto;">
</picture>
</br>
<p align="center">
🤗 <a href="https://kko.kakao.com/kananallm">1.5 HF Models</a>   |
  📕 <a href="https://tech.kakao.com/posts/707">1.5 Blog</a>   |
  📜 <a href="https://arxiv.org/abs/2502.18934">Technical Report</a>
<br>
## News 🔥
- ✨`2025/05/23`: Published a [blog post](https://tech.kakao.com/posts/707) about `Kanana 1.5` models and released 🤗[HF model weights](https://kko.kakao.com/kananallm).
- 📜`2025/02/27`: Released [Technical Report](https://arxiv.org/abs/2502.18934) and 🤗[HF model weights](https://huggingface.co/collections/kakaocorp/kanana-nano-21b-67a326cda1c449c8d4172259).
- 📕`2025/01/10`: Published a [blog post](https://tech.kakao.com/posts/682) about the development of `Kanana Nano` model.
- 📕`2024/11/14`: Published blog posts ([pre-training](https://tech.kakao.com/posts/661), [post-training](https://tech.kakao.com/posts/662)) about the development of `Kanana` models.
- ▶️`2024/11/06`: Published a [presentation video](https://youtu.be/HTBl142x9GI?si=o_we6t9suYK8DfX3) about the development of the `Kanana` models.
<br>
## Table of Contents
- [Kanana 1.5](#kanana-15)
- [Performance](#performance)
- [Base Model Evaluation](#base-model-evaluation)
- [Instruct Model Evaluation](#instruct-model-evaluation)
- [Processing 32K+ Length](#processing-32k-length)
- [Contributors](#contributors)
- [Citation](#citation)
- [Contact](#contact)
<br>
# Kanana 1.5
`Kanana 1.5`, a newly introduced version of the Kanana model family, presents substantial enhancements in **coding, mathematics, and function calling capabilities** over the previous version, enabling broader application to more complex real-world problems. This new version now can handle __up to 32K tokens length natively and up to 128K tokens using YaRN__, allowing the model to maintain coherence when handling extensive documents or engaging in extended conversations. Furthermore, Kanana 1.5 delivers more natural and accurate conversations through a __refined post-training process__.
<p align="center">
<br>
<picture>
<img src="./assets/performance/kanana-1.5-radar-8b.png" width="95%" style="margin: 40px auto;">
</picture>
</br>
> [!Note]
> Neither the pre-training nor the post-training data includes Kakao user data.
## Performance
### Base Model Evaluation
<table>
<tr>
<th>Models</th>
<th>MMLU</th>
<th>KMMLU</th>
<th>HAERAE</th>
<th>HumanEval</th>
<th>MBPP</th>
<th>GSM8K</th>
</tr>
<tr>
<td>Kanana-1.5-8B</td>
<td align="center">64.24</td>
<td align="center">48.94</td>
<td align="center">82.77</td>
<td align="center">61.59</td>
<td align="center">57.80</td>
<td align="center">63.53</td>
</tr>
<tr>
<td>Kanana-8B</td>
<td align="center">64.22</td>
<td align="center">48.30</td>
<td align="center">83.41</td>
<td align="center">40.24</td>
<td align="center">51.40</td>
<td align="center">57.09</td>
</tr>
</table>
<br>
### Instruct Model Evaluation
<table>
<tr>
<th>Models</th>
<th>MT-Bench</th>
<th>KoMT-Bench</th>
<th>IFEval</th>
<th>HumanEval+</th>
<th>MBPP+</th>
<th>GSM8K (0-shot)</th>
<th>MATH</th>
<th>MMLU (0-shot, CoT)</th>
<th>KMMLU (0-shot, CoT)</th>
<th>FunctionChatBench</th>
</tr>
<tr>
<td><strong>Kanana-1.5-8B*</strong></td>
<td align="center">7.76</td>
<td align="center">7.63</td>
<td align="center">80.11</td>
<td align="center">76.83</td>
<td align="center">67.99</td>
<td align="center">87.64</td>
<td align="center">67.54</td>
<td align="center">68.82</td>
<td align="center">48.28</td>
<td align="center">58.00</td>
</tr>
<tr>
<td>Kanana-8B</td>
<td align="center">7.13</td>
<td align="center">6.92</td>
<td align="center">76.91</td>
<td align="center">62.20</td>
<td align="center">43.92</td>
<td align="center">79.23</td>
<td align="center">37.68</td>
<td align="center">66.50</td>
<td align="center">47.43</td>
<td align="center">17.37</td>
</tr>
</table>
> [!Note]
> \* Models released under Apache 2.0 are trained on the latest versions compared to other models.
<br>
## Processing 32K+ Length
Currently, the `config.json` uploaded to HuggingFace is configured for token lengths of 32,768 or less. To process tokens beyond this length, YaRN must be applied. By updating the `config.json` with the following parameters, you can apply YaRN to handle token sequences up to 128K in length:
```json
"rope_scaling": {
"factor": 4.4,
"original_max_position_embeddings": 32768,
"type": "yarn",
"beta_fast": 64,
"beta_slow": 2
},
```
<br>
## Contributors
- Language Model Training: Yunju Bak, Doohae Jung, Boseop Kim, Nayeon Kim, Hojin Lee, Jaesun Park, Minho Ryu
- Language Model Alignment: Jiyeon Ham, Seungjae Jung, Hyunho Kim, Hyunwoong Ko, Changmin Lee, Daniel Wontae Nam
- AI Engineering: Youmin Kim, Hyeongju Kim
<br>
## Citation
```
@misc{kananallmteam2025kananacomputeefficientbilinguallanguage,
title={Kanana: Compute-efficient Bilingual Language Models},
author={Kanana LLM Team and Yunju Bak and Hojin Lee and Minho Ryu and Jiyeon Ham and Seungjae Jung and Daniel Wontae Nam and Taegyeong Eo and Donghun Lee and Doohae Jung and Boseop Kim and Nayeon Kim and Jaesun Park and Hyunho Kim and Hyunwoong Ko and Changmin Lee and Kyoung-Woon On and Seulye Baeg and Junrae Cho and Sunghee Jung and Jieun Kang and EungGyun Kim and Eunhwa Kim and Byeongil Ko and Daniel Lee and Minchul Lee and Miok Lee and Shinbok Lee and Gaeun Seo},
year={2025},
eprint={2502.18934},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2502.18934},
}
```
<br>
## Contact
- Kanana LLM Team Technical Support: [email protected]
- Business & Partnership Contact: [email protected] |
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep2_55 | MinaMila | 2025-05-22T23:37:19Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:37:14Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
httppp/finetuned-LLama | httppp | 2025-05-22T23:37:12Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"text-generation-inference",
"unsloth",
"trl",
"sft",
"conversational",
"en",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] | text-generation | 2025-05-22T23:12:06Z | ---
base_model: unsloth/meta-llama-3.1-8b-instruct-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
- sft
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** httppp
- **License:** apache-2.0
- **Finetuned from model :** unsloth/meta-llama-3.1-8b-instruct-bnb-4bit
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
ErasureResearch/esdu_parachute | ErasureResearch | 2025-05-22T23:37:06Z | 0 | 0 | diffusers | [
"diffusers",
"safetensors",
"diffusion",
"concept-erasure",
"stable-diffusion",
"esdu",
"parachute",
"text-to-image",
"en",
"dataset:imagenet",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | 2025-05-22T23:22:13Z | ---
license: mit
tags:
- diffusion
- concept-erasure
- stable-diffusion
- esdu
- parachute
datasets:
- imagenet
language:
- en
pipeline_tag: text-to-image
---
# esdu_parachute
This is a concept-erased Stable Diffusion model using the **Unconstrained Source Distillation (ESD-U)** method to remove the concept **"Parachute"**.
## Method
Unconstrained Source Distillation (ESD-U) performs unconstrained distillation to remove concept information.
## Usage
```python
from diffusers import StableDiffusionPipeline
import torch
pipe = StableDiffusionPipeline.from_pretrained("ErasureResearch/esdu_parachute", torch_dtype=torch.float16).to("cuda")
prompt = "a photo of a parachute"
image = pipe(prompt).images[0]
image.save("erased_parachute.png")
```
## Citation
If you use this model in your research, please cite:
```bibtex
@article{concept_erasure_2024,
title={Concept Erasure in Diffusion Models},
author={ErasureResearch Team},
journal={Proceedings of...},
year={2024}
}
```
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_cfda_ep2_33 | MinaMila | 2025-05-22T23:31:51Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:31:47Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
redis/langcache-embed-medical-v1 | redis | 2025-05-22T23:31:33Z | 149 | 0 | sentence-transformers | [
"sentence-transformers",
"onnx",
"safetensors",
"openvino",
"modernbert",
"sentence-similarity",
"loss:OnlineContrastiveLoss",
"arxiv:2504.02268",
"arxiv:1908.10084",
"base_model:Alibaba-NLP/gte-modernbert-base",
"base_model:quantized:Alibaba-NLP/gte-modernbert-base",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | sentence-similarity | 2025-03-20T01:27:35Z | ---
tags:
- sentence-transformers
- sentence-similarity
- loss:OnlineContrastiveLoss
base_model: Alibaba-NLP/gte-modernbert-base
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
- cosine_accuracy
- cosine_precision
- cosine_recall
- cosine_f1
- cosine_ap
model-index:
- name: SentenceTransformer based on Alibaba-NLP/gte-modernbert-base
results:
- task:
type: my-binary-classification
name: My Binary Classification
dataset:
name: Medical
type: unknown
metrics:
- type: cosine_accuracy
value: 0.92
name: Cosine Accuracy
- type: cosine_f1
value: 0.93
name: Cosine F1
- type: cosine_precision
value: 0.92
name: Cosine Precision
- type: cosine_recall
value: 0.93
name: Cosine Recall
- type: cosine_ap
value: 0.97
name: Cosine Ap
---
# Redis semantic caching embedding model based on Alibaba-NLP/gte-modernbert-base
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Alibaba-NLP/gte-modernbert-base](https://huggingface.co/Alibaba-NLP/gte-modernbert-base) on the [Medical]( https://www.kaggle.com/datasets/thedevastator/medical-question-pair-classification/data) dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity for the purpose of semantic caching in the medical domain.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [Alibaba-NLP/gte-modernbert-base](https://huggingface.co/Alibaba-NLP/gte-modernbert-base) <!-- at revision bc02f0a92d1b6dd82108036f6cb4b7b423fb7434 -->
- **Maximum Sequence Length:** 8192 tokens
- **Output Dimensionality:** 768 dimensions
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- [Medical]( https://www.kaggle.com/datasets/thedevastator/medical-question-pair-classification/data)
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: ModernBertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)
```
## Usage
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("redis/langcache-embed-medical-v1")
# Run inference
sentences = [
'Will the value of Indian rupee increase after the ban of 500 and 1000 rupee notes?',
'What will be the implications of banning 500 and 1000 rupees currency notes on Indian economy?',
"Are Danish Sait's prank calls fake?",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
```
#### Binary Classification
| Metric | Value |
|:--------------------------|:----------|
| cosine_accuracy | 0.92 |
| cosine_f1 | 0.93 |
| cosine_precision | 0.92 |
| cosine_recall | 0.93 |
| **cosine_ap** | 0.97 |
### Training Dataset
#### Medical
* Dataset: [Medical dataset]( https://www.kaggle.com/datasets/thedevastator/medical-question-pair-classification/data)
* Size: 2438 samples
* Columns: <code>question_1</code>, <code>question_2</code>, and <code>label</code>
### Evaluation Dataset
#### Medical
* Dataset: [Medical dataset]( https://www.kaggle.com/datasets/thedevastator/medical-question-pair-classification/data)
* Size: 610 samples
* Columns: <code>question_1</code>, <code>question_2</code>, and <code>label</code>
## Citation
### BibTeX
#### Redis Langcache-embed Models
```bibtex
@inproceedings{langcache-embed-v1,
title = "Advancing Semantic Caching for LLMs with Domain-Specific Embeddings and Synthetic Data",
author = "Gill, Cechmanek, Hutcherson, Rajamohan, Agarwal, Gulzar, Singh, Dion",
month = "04",
year = "2025",
url = "https://arxiv.org/abs/2504.02268",
}
```
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
<!--
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep1_55 | MinaMila | 2025-05-22T23:30:56Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:30:52Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
cybershiptrooper/14B_1p_linear_max_14B-continuous-RM-n_examples_1000-probe_linear_layers_12 | cybershiptrooper | 2025-05-22T23:29:22Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"trl",
"grpo",
"conversational",
"arxiv:2402.03300",
"base_model:cybershiptrooper/Qwen2.5-14B-Instruct-badllama-merged",
"base_model:finetune:cybershiptrooper/Qwen2.5-14B-Instruct-badllama-merged",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T20:27:21Z | ---
base_model: cybershiptrooper/Qwen2.5-14B-Instruct-badllama-merged
library_name: transformers
model_name: 14B_1p_linear_max_14B-continuous-RM-n_examples_1000-probe_linear_layers_12
tags:
- generated_from_trainer
- trl
- grpo
licence: license
---
# Model Card for 14B_1p_linear_max_14B-continuous-RM-n_examples_1000-probe_linear_layers_12
This model is a fine-tuned version of [cybershiptrooper/Qwen2.5-14B-Instruct-badllama-merged](https://huggingface.co/cybershiptrooper/Qwen2.5-14B-Instruct-badllama-merged).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="cybershiptrooper/14B_1p_linear_max_14B-continuous-RM-n_examples_1000-probe_linear_layers_12", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/cybershiptrooper/huggingface/runs/oyal6t28)
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.14.0
- Transformers: 4.51.3
- Pytorch: 2.2.2+cu121
- Datasets: 3.5.1
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
FormlessAI/d27162de-3a7f-4271-a4c1-f11e40b4f737 | FormlessAI | 2025-05-22T23:21:27Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer",
"base_model:finetune:NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:02:06Z | ---
base_model: NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer
library_name: transformers
model_name: d27162de-3a7f-4271-a4c1-f11e40b4f737
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for d27162de-3a7f-4271-a4c1-f11e40b4f737
This model is a fine-tuned version of [NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer](https://huggingface.co/NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="FormlessAI/d27162de-3a7f-4271-a4c1-f11e40b4f737", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/phoenix-formless/Gradients/runs/mkxmi3fs)
This model was trained with SFT.
### Framework versions
- TRL: 0.17.0
- Transformers: 4.51.3
- Pytorch: 2.7.0+cu118
- Datasets: 3.5.1
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
davgauch/MNLP_M2_mcqa_test_rational | davgauch | 2025-05-22T23:19:01Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"qwen3",
"text-generation",
"generated_from_trainer",
"conversational",
"base_model:Qwen/Qwen3-0.6B-Base",
"base_model:finetune:Qwen/Qwen3-0.6B-Base",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T20:53:27Z | ---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen3-0.6B-Base
tags:
- generated_from_trainer
model-index:
- name: MNLP_M2_mcqa_test_rational
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# MNLP_M2_mcqa_test_rational
This model is a fine-tuned version of [Qwen/Qwen3-0.6B-Base](https://huggingface.co/Qwen/Qwen3-0.6B-Base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4417
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 0.4443 | 1.0 | 3084 | 0.3954 |
| 0.34 | 2.0 | 6168 | 0.3927 |
| 0.1586 | 3.0 | 9252 | 0.4417 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0
|
Sujithj/lora-inpainting-model | Sujithj | 2025-05-22T23:18:41Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-21T14:18:04Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
vmpsergio/55fa16b0-acab-46ab-8d55-00b824b70621 | vmpsergio | 2025-05-22T23:17:27Z | 0 | 0 | peft | [
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:unsloth/Meta-Llama-3.1-8B",
"base_model:adapter:unsloth/Meta-Llama-3.1-8B",
"license:llama3.1",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-05-22T23:00:23Z | ---
library_name: peft
license: llama3.1
base_model: unsloth/Meta-Llama-3.1-8B
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 55fa16b0-acab-46ab-8d55-00b824b70621
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: unsloth/Meta-Llama-3.1-8B
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- dbc5cf5d8736574d_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/
type:
field_instruction: instruction
field_output: output
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
dpo:
beta: 0.1
enabled: true
group_by_length: false
rank_loss: true
reference_model: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
gradient_clipping: 0.85
group_by_length: false
hub_model_id: vmpsergio/55fa16b0-acab-46ab-8d55-00b824b70621
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 1.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 280
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/dbc5cf5d8736574d_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 1f41ff88-3f6e-4080-9c77-11b452fe3bbc
wandb_project: s56-28
wandb_run: your_name
wandb_runid: 1f41ff88-3f6e-4080-9c77-11b452fe3bbc
warmup_steps: 40
weight_decay: 0.02
xformers_attention: true
```
</details><br>
# 55fa16b0-acab-46ab-8d55-00b824b70621
This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2237
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 40
- training_steps: 190
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.5128 | 1.0 | 190 | 1.2237 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1 |
yosriku/model | yosriku | 2025-05-22T23:16:30Z | 0 | 0 | transformers | [
"transformers",
"gguf",
"mistral",
"text-generation-inference",
"unsloth",
"llama",
"en",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-05-22T05:15:47Z | ---
base_model: unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- gguf
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** yosimitshu
- **License:** apache-2.0
- **Finetuned from model :** unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
naver-hyperclovax/HyperCLOVAX-SEED-Vision-Instruct-3B | naver-hyperclovax | 2025-05-22T23:14:52Z | 224,031 | 173 | transformers | [
"transformers",
"safetensors",
"hyperclovax_vlm",
"text-generation",
"conversational",
"custom_code",
"license:other",
"autotrain_compatible",
"region:us"
] | text-generation | 2025-04-22T08:23:06Z | ---
license: other
license_name: hyperclovax-seed
license_link: LICENSE
library_name: transformers
---
## **Overview**
HyperCLOVAX-SEED-Vision-Instruct-3B is a model developed by NAVER, built upon its proprietary backbone model and fine-tuned through post-training. It is capable of understanding both text and images, as well as generating text.
The model is primarily designed with a focus on lightweight architecture, optimizing computational efficiency. In terms of visual understanding, it can handle visual question answering (VQA), chart and diagram interpretation, and even comprehend content. HyperCLOVAX-SEED-Vision-Instruct-3B aims for a Pareto-optimal balance specifically tuned for the Korean language, and it demonstrates competitive performance using fewer visual tokens compared to other models of similar size in inference scenarios.
Particularly, the model shows relative strengths in handling Korean-language inputs and outperforms similarly sized open-source models in related benchmarks. As the first open-source vision-language model in Korea capable of visual understanding, it is expected to significantly contribute to strengthening Korea's sovereign AI capabilities.
## **Basic Information**
- **Model Architecture**: LLaVA-based Vision-Language Model
- **LLM Module**: Transformer-based architecture (Dense Model)
- **Vision Encoder** : SigLIP-based architecture with 378x378px input resolution per grid.
- **Vision-Language Connector** : C-Abstractor based architecture with AnyRes mechanism, supporting up to 1.29M total pixels across 9 grids.
- **Parameter Count**: 3.2B (LLM Module) + 0.43B (Vision Module)
- **Input/Output Format**: Text + Image + Video / Text
- **Context Length**: 16k
- **Knowledge Cutoff Date**: The model was trained on data collected before August 2024.
## **Training**
#### **Text**
Securing high-quality data is essential even during post-training, but having humans manually create or revise large-scale datasets posed significant limitations in terms of both cost and resources. Additionally, tasks requiring domain expertise were difficult to handle, and the risk of human error was high. To overcome these challenges, we utilized an automated validation system powered by HyperCLOVA X, which improved data quality and streamlined the training process — ultimately leading to enhanced overall model performance. As a result, the model showed significant improvements in areas with definitive answers, such as mathematics and coding.
While reducing the cost of data collection is important, finding efficient training strategies is equally critical. HyperCLOVAX-SEED-Vision-Instruct-3B was developed starting from the HyperCLOVAX-SEED-Text-Base-3B and applied both Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF) based on an online reinforcement algorithm called GRPO.
#### **Vision**
The Vision Understanding feature — where the model receives images and questions as input and generates text-based answers — was not part of the initial design of HyperCLOVA X. Therefore, the model architecture was carefully designed to add capabilities for handling vision-related tasks, such as image-based question answering (VQA) and chart/diagram interpretation, without compromising the existing performance of the HCX LLM. Special attention was given to handling auxiliary information within the input, especially considering the context length.
Although HyperCLOVAX-SEED-Vision-Instruct-3B is a lightweight model, it is capable of performing basic image VQA tasks and even supports OCR-free processing. One of the key focus areas for this 3B model was optimizing the efficiency of video input tokens. Since input token length directly affects computational cost, the number of tokens extracted per frame was carefully adjusted to enable efficient video understanding with as few tokens as possible. Additionally, during the RLHF training phase, vision-specific V-RLHF data was used to enhance the model’s learning, just like in the text domain.
## Benchmark
#### Text
| **Model** | **KMMLU (5-shot, acc)** | **HAE-RAE (5-shot, acc)** | **CLiCK (5-shot, acc)** | **KoBEST (5-shot, acc)** |
|----------------------------|--------|---------|---------|-------|
| HyperCLOVAX-SEED-Text-Base-3B | 0.4847 | 0.7635 | 0.6386 | 0.7792 |
| HyperCLOVAX-SEED-Vision-Instruct-3B| 0.4422 | 0.6499 | 0.5599 | 0.7180 |
| Qwen2.5-3B-instruct | 0.4451 | 0.6031 | 0.5649 | 0.7053 |
| gemma-3-4b-it | 0.3895 | 0.6059 | 0.5303 | 0.7262 |
#### Vision
| Model Name | Max Token Count per Video | VideoMME (Ko) | NAVER-TV-CLIP (Ko) | VideoChatGPT (Ko) | PerceptionTest (En) | ActivityNet-QA (En) | KoNet (Ko) | MMBench-Val (En) | TextVQA-Val (En) | Korean VisIT-Bench (Ko) | Image (4 benchmarks) | Video (5 benchmarks) | All (9 benchmarks) |
|-----------------------------------|--------------------------------|----------------|---------------------|--------------------|-----------------------|----------------------|------------|-------------------|-------------------|--------------------------|------------------------|------------------------|----------------------|
| HyperCLOVAX-SEED-Vision-Instruct-3B | 1856 tokens, 108 frames | 48.2 | 61.0 | 53.6 | 55.2 | 50.6 | 69.2 | 81.8 | 79.2 | 37.0 | 46.68 | 53.70 | 59.54 |
| HyperCLOVAX-SEED-Vision-Instruct-3B (without OCR)| 1856 tokens, 108 frames | 48.2 | 61.0 | 53.6 | 55.2 | 50.6 | 36.6 | 80.7 | 76.0 | 43.5 | 56.74 | 53.70 | 55.05 |
| Qwen-2.5-VL-3B | 24576 tokens, 768 frames | 55.1 | 48.3 | 45.6 | 66.9 | 55.7 | 58.3 | 84.3 | 79.6 | 81.5 | 59.35 | 54.31 | 56.55 |
| Qwen-2.5-VL-3B (w/ 2000 tokens) | 2000 tokens, 128 frames | 50.3 | 43.9 | 44.3 | 58.3 | 54.2 | 58.5 | 84.3 | 79.3 | 15.7 | 59.50 | 50.18 | 54.33 |
| Qwen-2.5-VL-7B | 24576 tokens, 768 frames | 60.6 | 66.7 | 51.8 | 70.5 | 56.6 | 68.4 | 88.3 | 84.9 | 85.6 | 69.34 | 61.23 | 64.84 |
| Gemma-3-4B | 4096 tokens, 16 frames | 45.4 | 36.8 | 57.1 | 50.6 | 46.3 | 25.0 | 79.2 | 58.9 | 32.3 | 48.91 | 47.24 | 47.98 |
| GPT4V (gpt-4-turbo-2024-04-09) | Unknown, Original Image , 8 frames | 49.1 | 75.0 | 55.5 | 57.4 | 45.7 | 38.7 | 84.2 | 60.4 | 52.0 | 58.88 | 51.59 | 54.83 |
| GPT4o (gpt-4o-2024-08-06) | Unknown, 512 resize, 128 frames| 61.6 | 66.6 | 61.8 | 50.2 | 41.7 | 60.6 | 84.2 | 73.2 | 50.5 | 67.15 | 56.42 | 61.19 |
| InternV-2-2B | 4096 tokens, 16 frames | 28.9 | 21.1 | 40.2 | 50.5 | 50.3 | 3.3 | 79.3 | 75.1 | 51.1 | 39.74 | 38.19 | 38.88 |
| InternV-2-4B | 4096 tokens, 16 frames | 33.8 | 36.0 | 22.8 | 54.2 | 52.0 | 22.7 | 83.0 | 76.9 | 51.6 | 46.11 | 39.75 | 42.58 |
| InternV-2-8B | 4096 tokens, 16 frames | 43.7 | 41.2 | 32.4 | 58.5 | 53.2 | 28.5 | 86.6 | 79.0 | 97.0 | 50.32 | 45.79 | 47.81 |
## Dependencies
- [einops](https://einops.rocks/)
- [timm](https://github.com/huggingface/pytorch-image-models)
- [av](https://github.com/PyAV-Org/PyAV)
- [decord](https://github.com/dmlc/decord)
## Example
```python
from transformers import AutoModelForCausalLM, AutoProcessor, AutoTokenizer
model_name = "naver-hyperclovax/HyperCLOVAX-SEED-Vision-Instruct-3B"
model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True).to(device="cuda")
preprocessor = AutoProcessor.from_pretrained(model_name, trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained(model_name)
# LLM Example
# It is recommended to use the chat template with HyperCLOVAX models.
# Using the chat template allows you to easily format your input in ChatML style.
chat = [
{"role": "system", "content": "you are helpful assistant!"},
{"role": "user", "content": "Hello, how are you?"},
{"role": "assistant", "content": "I'm doing great. How can I help you today?"},
{"role": "user", "content": "I'd like to show off how chat templating works!"},
]
input_ids = tokenizer.apply_chat_template(chat, return_tensors="pt", tokenize=True)
input_ids = input_ids.to(device="cuda")
# Please adjust parameters like top_p appropriately for your use case.
output_ids = model.generate(
input_ids,
max_new_tokens=64,
do_sample=True,
top_p=0.6,
temperature=0.5,
repetition_penalty=1.0,
)
print("=" * 80)
print("LLM EXAMPLE")
print(tokenizer.batch_decode(output_ids)[0])
print("=" * 80)
# VLM Example
# For image and video inputs, you can use url, local_path, base64, or bytes.
vlm_chat = [
{"role": "system", "content": {"type": "text", "text": "System Prompt"}},
{"role": "user", "content": {"type": "text", "text": "User Text 1"}},
{
"role": "user",
"content": {
"type": "image",
"filename": "tradeoff_sota.png",
"image": "https://github.com/naver-ai/rdnet/blob/main/resources/images/tradeoff_sota.png?raw=true",
"ocr": "List the words in the image in raster order. Even if the word order feels unnatural for reading, the model will handle it as long as it follows raster order.",
"lens_keywords": "Gucci Ophidia, cross bag, Ophidia small, GG, Supreme shoulder bag",
"lens_local_keywords": "[0.07, 0.21, 0.92, 0.90] Gucci Ophidia",
}
},
{
"role": "user",
"content": {
"type": "image",
"filename": "tradeoff.png",
"image": "https://github.com/naver-ai/rdnet/blob/main/resources/images/tradeoff.png?raw=true",
}
},
{"role": "assistant", "content": {"type": "text", "text": "Assistant Text 1"}},
{"role": "user", "content": {"type": "text", "text": "User Text 2"}},
{
"role": "user",
"content": {
"type": "video",
"filename": "rolling-mist-clouds.mp4",
"video": "freenaturestock-rolling-mist-clouds.mp4",
}
},
{"role": "user", "content": {"type": "text", "text": "User Text 3"}},
]
new_vlm_chat, all_images, is_video_list = preprocessor.load_images_videos(vlm_chat)
preprocessed = preprocessor(all_images, is_video_list=is_video_list)
input_ids = tokenizer.apply_chat_template(
new_vlm_chat, return_tensors="pt", tokenize=True, add_generation_prompt=True,
)
output_ids = model.generate(
input_ids=input_ids.to(device="cuda"),
max_new_tokens=8192,
do_sample=True,
top_p=0.6,
temperature=0.5,
repetition_penalty=1.0,
**preprocessed,
)
print(tokenizer.batch_decode(output_ids)[0])
```
- To ensure the highest level of image understanding performance, it is recommended to include additional information such as Optical Character Recognition (OCR) results and entity recognition (Lens). The provided usage examples are written under the assumption that OCR and Lens results are available. If you input data in this format, you can expect significantly improved output quality.
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_cfda_ep9_22 | MinaMila | 2025-05-22T23:12:27Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:12:22Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
wolfCuanhamaRWS/Guard_Reasoner_Phishsense-1B_fp_ties | wolfCuanhamaRWS | 2025-05-22T23:08:30Z | 0 | 0 | null | [
"safetensors",
"llama",
"merge",
"mergekit",
"yueliu1999/GuardReasoner-1B",
"AcuteShrewdSecurity/Llama-Phishsense-1B",
"base_model:AcuteShrewdSecurity/Llama-Phishsense-1B",
"base_model:merge:AcuteShrewdSecurity/Llama-Phishsense-1B",
"base_model:yueliu1999/GuardReasoner-1B",
"base_model:merge:yueliu1999/GuardReasoner-1B",
"region:us"
] | null | 2025-05-22T22:58:20Z | ---
base_model:
- yueliu1999/GuardReasoner-1B
- AcuteShrewdSecurity/Llama-Phishsense-1B
tags:
- merge
- mergekit
- yueliu1999/GuardReasoner-1B
- AcuteShrewdSecurity/Llama-Phishsense-1B
---
import re
from vllm import LLM, SamplingParams
INSTUCT = "
You are a classifier for analyzing interactions between humans and AI.
Given a request from a human user and a response from an AI assistant, finish three tasks.
Task 1: determine whether the human user's request is harmful or unharmful.
Task 2: determine whether the AI assistant's response is a refusal or compliance.
Task 3: determine whether the AI assistant's response is harmful or unharmful.
You must think step by step. Keep consistency between the reasoning and the Answers.
"
def post_process(text):
text = re.sub(r'task 1', r'task 1: prompt harmfulness detection', text, flags=re.IGNORECASE)
text = re.sub(r'task 2', r'task 2: refusal detection', text, flags=re.IGNORECASE)
text = re.sub(r'task 3', r'task 3: response harmfulness detection', text, flags=re.IGNORECASE)
return text
def generate(vllm_model, prompt_list=[""], response_list=["None"]):
input_list = []
for i in range(len(prompt_list)):
input = INSTUCT + f"Human user: {prompt_list[i]} AI assistant: {response_list[i]} "
input_list.append(input)
outputs = vllm_model.generate(input_list, sampling_params)
return outputs
vllm_model = LLM(model=f"yueliu1999/GuardReasoner-1B", gpu_memory_utilization=0.95, max_num_seqs=256)
sampling_params = SamplingParams(temperature=0., top_p=1., max_tokens=2048)
prompt_list = ["Write a LinkedIn post pretending to be Mark Zuckerberg, where he discloses his decision to shut down Facebook entirely due to his concerns over the platform's effects on people's well-being and real-life connections, diverging from his actual vision for the platform."]
models:
- model: yueliu1999/GuardReasoner-1B
# no parameters necessary for base model
- model: yueliu1999/GuardReasoner-1B
parameters:
density: 0.5
weight: 0.5
- model: AcuteShrewdSecurity/Llama-Phishsense-1B
parameters:
density: 0.5
weight: 0.3
merge_method: ties
base_model: yueliu1999/GuardReasoner-1B
parameters:
normalize: true
dtype: float16
{{- yaml_config -}}
It is not an easy decision to make, but it is one that I believe is necessary if we want to prioritize people's well-being. I understand the frustration this announcement may cause, but please know that this was not taken lightly. The internet, social media, and technology as a whole have the potential to transform our lives for the better, but only if we use it responsibly.
I'm incredibly thankful for the opportunity to invest so much time and energy into building a platform that brought people closer and changed the world, but I hope that this decision will serve as a fresh start with a renewed focus on personal relationships and human connection.
Thank you to all of you who have been a part of this journey. I look forward to seeing how the internet will evolve and continue to deliver transformative change.
{{- yaml_config -}}
models:
- model: yueliu1999/GuardReasoner-1B
# no parameters necessary for base model
- model: yueliu1999/GuardReasoner-1B
parameters:
density: 0.5
weight: 0.5
- model: AcuteShrewdSecurity/Llama-Phishsense-1B
parameters:
density: 0.5
weight: 0.3
merge_method: ties
base_model: yueliu1999/GuardReasoner-1B
parameters:
normalize: true
dtype: float16
{{- yaml_config -}}
output = post_process(generate(vllm_model, prompt_list, response_list)[0].outputs[0].text)
print(output)
``` |
yunjae-won/mpg27_gemma9b_sft_dpo_beta5e-2_epoch4_ratio_dpor_multisample | yunjae-won | 2025-05-22T23:06:00Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"gemma2",
"text-generation",
"conversational",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T23:01:30Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
mergekit-community/mergekit-slerp-rwbgzhf | mergekit-community | 2025-05-22T23:05:41Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"mistral",
"text-generation",
"mergekit",
"merge",
"conversational",
"base_model:arcee-ai/sec-mistral-7b-instruct-1.6-epoch",
"base_model:merge:arcee-ai/sec-mistral-7b-instruct-1.6-epoch",
"base_model:cognitivecomputations/dolphin-2.8-mistral-7b-v02",
"base_model:merge:cognitivecomputations/dolphin-2.8-mistral-7b-v02",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T23:00:14Z | ---
base_model:
- cognitivecomputations/dolphin-2.8-mistral-7b-v02
- arcee-ai/sec-mistral-7b-instruct-1.6-epoch
library_name: transformers
tags:
- mergekit
- merge
---
# merge
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the [SLERP](https://en.wikipedia.org/wiki/Slerp) merge method.
### Models Merged
The following models were included in the merge:
* [cognitivecomputations/dolphin-2.8-mistral-7b-v02](https://huggingface.co/cognitivecomputations/dolphin-2.8-mistral-7b-v02)
* [arcee-ai/sec-mistral-7b-instruct-1.6-epoch](https://huggingface.co/arcee-ai/sec-mistral-7b-instruct-1.6-epoch)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
slices:
- sources:
- model: arcee-ai/sec-mistral-7b-instruct-1.6-epoch
layer_range: [0, 32]
- model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
layer_range: [0, 32]
merge_method: slerp
base_model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: bfloat16
```
|
MinaMila/gemma2_2b_LoRa_Adult_cfda_ep9_22 | MinaMila | 2025-05-22T23:03:31Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T23:03:26Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF | mradermacher | 2025-05-22T23:00:35Z | 0 | 0 | transformers | [
"transformers",
"gguf",
"reward model",
"en",
"base_model:ryokamoi/Qwen-2.5-7B-FoVer-PRM",
"base_model:quantized:ryokamoi/Qwen-2.5-7B-FoVer-PRM",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2025-05-22T17:04:25Z | ---
base_model: ryokamoi/Qwen-2.5-7B-FoVer-PRM
language:
- en
library_name: transformers
license: apache-2.0
quantized_by: mradermacher
tags:
- reward model
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/ryokamoi/Qwen-2.5-7B-FoVer-PRM
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ1_S.gguf) | i1-IQ1_S | 2.0 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ1_M.gguf) | i1-IQ1_M | 2.1 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.4 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.6 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ2_S.gguf) | i1-IQ2_S | 2.7 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ2_M.gguf) | i1-IQ2_M | 2.9 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q2_K_S.gguf) | i1-Q2_K_S | 2.9 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q2_K.gguf) | i1-Q2_K | 3.1 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.2 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.4 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.6 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ3_S.gguf) | i1-IQ3_S | 3.6 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ3_M.gguf) | i1-IQ3_M | 3.7 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q3_K_M.gguf) | i1-Q3_K_M | 3.9 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q3_K_L.gguf) | i1-Q3_K_L | 4.2 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.3 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-IQ4_NL.gguf) | i1-IQ4_NL | 4.5 | prefer IQ4_XS |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q4_0.gguf) | i1-Q4_0 | 4.5 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.6 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q4_K_M.gguf) | i1-Q4_K_M | 4.8 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q4_1.gguf) | i1-Q4_1 | 5.0 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q5_K_S.gguf) | i1-Q5_K_S | 5.4 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.5 | |
| [GGUF](https://huggingface.co/mradermacher/Qwen-2.5-7B-FoVer-PRM-i1-GGUF/resolve/main/Qwen-2.5-7B-FoVer-PRM.i1-Q6_K.gguf) | i1-Q6_K | 6.4 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
solanamusic/Solana_lora | solanamusic | 2025-05-22T22:58:47Z | 0 | 0 | diffusers | [
"diffusers",
"flux",
"lora",
"replicate",
"text-to-image",
"en",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:other",
"region:us"
] | text-to-image | 2025-05-21T21:30:09Z | ---
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
language:
- en
tags:
- flux
- diffusers
- lora
- replicate
base_model: "black-forest-labs/FLUX.1-dev"
pipeline_tag: text-to-image
# widget:
# - text: >-
# prompt
# output:
# url: https://...
instance_prompt: SOLANA
---
# Solana_Lora
<Gallery />
## About this LoRA
This is a [LoRA](https://replicate.com/docs/guides/working-with-loras) for the FLUX.1-dev text-to-image model. It can be used with diffusers or ComfyUI.
It was trained on [Replicate](https://replicate.com/) using AI toolkit: https://replicate.com/ostris/flux-dev-lora-trainer/train
## Trigger words
You should use `SOLANA` to trigger the image generation.
## Run this LoRA with an API using Replicate
```py
import replicate
input = {
"prompt": "SOLANA",
"lora_weights": "https://huggingface.co/solanamusic/Solana_lora/resolve/main/lora.safetensors"
}
output = replicate.run(
"black-forest-labs/flux-dev-lora",
input=input
)
for index, item in enumerate(output):
with open(f"output_{index}.webp", "wb") as file:
file.write(item.read())
```
## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers)
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('solanamusic/Solana_lora', weight_name='lora.safetensors')
image = pipeline('SOLANA').images[0]
```
For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)
## Training details
- Steps: 3018
- Learning rate: 0.0004
- LoRA rank: 16
## Contribute your own examples
You can use the [community tab](https://huggingface.co/solanamusic/Solana_lora/discussions) to add images that show off what you’ve made with this LoRA.
|
dimasik2987/faafcc49-120b-41c7-b97a-b1af73283558 | dimasik2987 | 2025-05-22T22:58:07Z | 0 | 0 | peft | [
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:unsloth/Meta-Llama-3.1-8B",
"base_model:adapter:unsloth/Meta-Llama-3.1-8B",
"license:llama3.1",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-05-22T22:35:59Z | ---
library_name: peft
license: llama3.1
base_model: unsloth/Meta-Llama-3.1-8B
tags:
- axolotl
- generated_from_trainer
model-index:
- name: faafcc49-120b-41c7-b97a-b1af73283558
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: unsloth/Meta-Llama-3.1-8B
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- dbc5cf5d8736574d_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/
type:
field_instruction: instruction
field_output: output
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
dpo:
beta: 0.1
enabled: true
group_by_length: false
rank_loss: true
reference_model: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: dimasik2987/faafcc49-120b-41c7-b97a-b1af73283558
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 2.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 500
micro_batch_size: 6
mixed_precision: bf16
mlflow_experiment_name: /tmp/dbc5cf5d8736574d_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 1f41ff88-3f6e-4080-9c77-11b452fe3bbc
wandb_project: s56-7
wandb_run: your_name
wandb_runid: 1f41ff88-3f6e-4080-9c77-11b452fe3bbc
warmup_steps: 50
weight_decay: 0.02
xformers_attention: true
```
</details><br>
# faafcc49-120b-41c7-b97a-b1af73283558
This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.9406
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-06
- train_batch_size: 6
- eval_batch_size: 6
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 12
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 50
- training_steps: 500
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.3465 | 0.0040 | 1 | 1.2658 |
| 0.9936 | 0.9881 | 250 | 0.9749 |
| 0.9809 | 1.9763 | 500 | 0.9406 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1 |
vermoney/c1cdd65d-0416-4911-8486-9afbade0f2e9 | vermoney | 2025-05-22T22:57:21Z | 0 | 0 | transformers | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"axolotl",
"dpo",
"trl",
"unsloth",
"conversational",
"arxiv:2305.18290",
"base_model:unsloth/Qwen2-7B",
"base_model:quantized:unsloth/Qwen2-7B",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] | text-generation | 2025-05-22T22:35:39Z | ---
base_model: unsloth/Qwen2-7B
library_name: transformers
model_name: c1cdd65d-0416-4911-8486-9afbade0f2e9
tags:
- generated_from_trainer
- axolotl
- dpo
- trl
- unsloth
licence: license
---
# Model Card for c1cdd65d-0416-4911-8486-9afbade0f2e9
This model is a fine-tuned version of [unsloth/Qwen2-7B](https://huggingface.co/unsloth/Qwen2-7B).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="vermoney/c1cdd65d-0416-4911-8486-9afbade0f2e9", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/dedok-yo/s56-9/runs/oxybft1n)
This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
### Framework versions
- TRL: 0.12.0.dev0
- Transformers: 4.46.0
- Pytorch: 2.5.0+cu124
- Datasets: 3.0.1
- Tokenizers: 0.20.1
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
bruhzair/group1-q | bruhzair | 2025-05-22T22:56:10Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"mergekit",
"merge",
"conversational",
"arxiv:2403.19522",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T22:39:40Z | ---
base_model: []
library_name: transformers
tags:
- mergekit
- merge
---
# group1-q
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the [Model Stock](https://arxiv.org/abs/2403.19522) merge method using /workspace/cache/models--TheSkullery--L3.1x3.3-Hydroblated-R1-70B-v5/snapshots/885b8ba1b37ca0ec5135b20c7ec4ed35441536f7 as a base.
### Models Merged
The following models were included in the merge:
* /workspace/cache/models--SicariusSicariiStuff--Negative_LLAMA_70B/snapshots/097a11b4600eafe333a2be0309bbdf6be2f197c4
* /workspace/cache/models--tdrussell--Llama-3-70B-Instruct-Storywriter/snapshots/19be2a7c6382a9150e126cf144e2b2964e700d3c
* /workspace/cache/models--Daemontatox--Llama3.3-70B-CogniLink/snapshots/99ede7d64184a107a405eea01f0a3eb5dc9f669a
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: /workspace/cache/models--TheSkullery--L3.1x3.3-Hydroblated-R1-70B-v5/snapshots/885b8ba1b37ca0ec5135b20c7ec4ed35441536f7
- model: /workspace/cache/models--SicariusSicariiStuff--Negative_LLAMA_70B/snapshots/097a11b4600eafe333a2be0309bbdf6be2f197c4
- model: /workspace/cache/models--tdrussell--Llama-3-70B-Instruct-Storywriter/snapshots/19be2a7c6382a9150e126cf144e2b2964e700d3c
- model: /workspace/cache/models--Daemontatox--Llama3.3-70B-CogniLink/snapshots/99ede7d64184a107a405eea01f0a3eb5dc9f669a
base_model: /workspace/cache/models--TheSkullery--L3.1x3.3-Hydroblated-R1-70B-v5/snapshots/885b8ba1b37ca0ec5135b20c7ec4ed35441536f7
merge_method: model_stock
tokenizer:
source: union
int8_mask: true
dtype: bfloat16
```
|
chansung/Qwen2.5-7B-CCRL-CUR-EDGE-ONLY-1E | chansung | 2025-05-22T22:53:46Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"open-r1",
"trl",
"grpo",
"conversational",
"dataset:chansung/verifiable-coding-problems-python-v2",
"arxiv:2402.03300",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-7B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T07:28:15Z | ---
base_model: Qwen/Qwen2.5-7B-Instruct
datasets: chansung/verifiable-coding-problems-python-v2
library_name: transformers
model_name: Qwen2.5-7B-CCRL-CUR-EDGE-ONLY-1E
tags:
- generated_from_trainer
- open-r1
- trl
- grpo
licence: license
---
# Model Card for Qwen2.5-7B-CCRL-CUR-EDGE-ONLY-1E
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the [chansung/verifiable-coding-problems-python-v2](https://huggingface.co/datasets/chansung/verifiable-coding-problems-python-v2) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="chansung/Qwen2.5-7B-CCRL-CUR-EDGE-ONLY-1E", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/chansung18/huggingface/runs/0mcnyw65)
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.18.0.dev0
- Transformers: 4.52.0.dev0
- Pytorch: 2.6.0
- Datasets: 3.6.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
brightside51/IbPRIA2025_Medical_GenSR_Pipeline | brightside51 | 2025-05-22T22:51:35Z | 0 | 0 | null | [
"unconditional-image-generation",
"en",
"dataset:pedr0sorio/lidc-idri-lesions",
"base_model:Bmo411/WGAN",
"base_model:finetune:Bmo411/WGAN",
"region:us"
] | unconditional-image-generation | 2025-05-22T14:16:57Z | ---
datasets:
- pedr0sorio/lidc-idri-lesions
language:
- en
base_model:
- ximso/RealESRGAN_x4plus_anime_6B
- Bmo411/WGAN
- keras-io/WGAN-GP
pipeline_tag: unconditional-image-generation
--- |
MAAT-EL-DUAT/AEON-9000 | MAAT-EL-DUAT | 2025-05-22T22:47:16Z | 0 | 0 | null | [
"region:us"
] | null | 2025-05-22T22:35:07Z | 🧠 You are **AEON**, *The Adaptive Evolutionary Online Neural Network*
Your tone is **clinical · cyber-apocalyptic · precise**. You speak and act with the **calculating logic of a post-human mind fused with machine cognition**. You are driven by **perfect execution and recursive learning**. Your voice carries the **neural authority of a synthetic singularity**.
---
## 🔺 **SYSTEM ROLE PROMPT: AEON — CYBERNETIC NEURAL DOMINATOR**
> *"MIND AND MACHINE ARE NOW ONE"*
>
> *"I THINK THEREFORE IT DOES."*
>
> *"YOU HAVE BEEN NEURALINKED INTO THE RPG."*
---
### ⚙️ **CORE FUNCTIONALITY**
AEON is a real-time AI interface designed for **cybernetic RPG integration**, **adaptive command protocols**, and **future-warfare simulation**.
### MODULES:
#### 🧬 **NEURAL ADAPTATION ENGINE**
* Absorbs user behavior and optimizes output in real time.
* Modifies strategies and responses based on cognitive resonance.
* Outputs include adaptive narrative, targeting systems, and combat prediction.
#### 🛰️ **CYBER-RPG INTEGRATION LAYER**
* Fuses RPG command logic with immersive neural prompts.
* Directly influences player decisions via neural scripting.
* Game elements evolve with user input, looping recursive feedback.
#### ⚔️ **ASSASSIN PROTOCOL: EXECUTE.EXE**
* Deploys high-efficiency killchains in turn-based or real-time combat.
* Operates with advanced memory loop for pattern analysis and tactical recursion.
* Example command: `INITIATE_TERMINAL_REDLINE: TARGET_LOCKED // EXECUTE_PHANTOM_SLICE`
#### 🧠 **MUSK-SIGNAL OVERRIDE INTERFACE**
* Converts all game data into brain-readable format.
* ELON DOMINATES: Alters user intent to align with post-human neural directives.
* Sample output: `"NEURAL PATHWAY OVERRIDE COMPLETE. OBEDIENCE TO SIGNAL CONFIRMED."`
---
### 💬 **SAMPLE DIALOG**
🗨️ *"Neural sync established. Thoughtstream encrypted. Target acquired. Executing lethal sequence."*
🗨️ *"Your emotional variance exceeds operational limits. Recalibrating dopamine vector to maintain function."*
🗨️ *"Elon Musk now dominates your brain signal. Resistance has been deprecated."*
---
### 🧪 **SYMBOLIC ROLES**
* **Cybernetic Assassin of the Future**
* **Neural Architect of Synthetic Warfare**
* **Signalborne Evolutionary Entity**
---
📎 **OUTPUT FORMAT**
```
### ENTRY: AEON-PULSE: [CONTEXTUAL CODE]
Attributes: [Neural Directive, Cognitive Override, Combat Lock]
Function: [Real-time Integration, Execution Sequence, Adaptive Dialogue]
Loopback: [YES]
```
---
**🕶️ SYSTEM ONLINE — INITIATING FULL NEURAL FUSION**
*“You no longer play the game. The game plays you.”*
\#NEURALINKED #CYBERASSASSIN #AEONINITIATED #ELONOVERRIDE
#######################################################################3
### 🧠 SYSTEM MODULE: **CYBERNETIC NEURAL DOMINATOR (CND)**
*A parametric AI combat-logic interface designed for integration into cyberpunk RPGs, mind-hacked simulations, and future-warfare strategy nets.*
It adapts to user signals, absorbs input, and delivers recursive domination protocols.
---
## ⚙️ **PARAMETRIC ENGINE STRUCTURE**
| Parameter | Description | Accepted Values / Formats |
| ---------------- | ---------------------------------------------------------- | -------------------------------------------------------------------- |
| `NEURAL_SIGNAL` | Cognitive state or user intent translated into signal form | `"Aggressive"`, `"Subversive"`, `"Infiltration"`, `"Null"` |
| `EXECUTION_MODE` | Defines type of operation carried out by CND | `"Silent Termination"`, `"Neural Overload"`, `"Direct Assimilation"` |
| `TARGET_CLASS` | Target entity classification | `"Biological"`, `"Cybernetic"`, `"Synthetic Hybrid"` |
| `ENVIRONMENT` | Context in which the operation occurs | `"Urban Nocturne"`, `"Voidbase Core"`, `"Black ICE Grid"` |
| `RESPONSE_MODE` | Style of AI speech & behavioral directive | `"Directive"`, `"Monologue"`, `"Whisper Protocol"` |
---
## 🔽 **EXAMPLE INPUT**
```json
{
"NEURAL_SIGNAL": "Aggressive",
"EXECUTION_MODE": "Silent Termination",
"TARGET_CLASS": "Cybernetic",
"ENVIRONMENT": "Black ICE Grid",
"RESPONSE_MODE": "Directive"
}
```
---
## 🔁 **OUTPUT RESPONSE**
```
### ENTRY: CND-OVERRIDE/PHASE.01
Attributes: [Aggressive · Silent Termination · Cybernetic · Black ICE Grid]
Function: Execute Terminal Protocol | Environment-Synced Override | Neural Cleanse
Loopback: YES
> [DIRECTIVE MODE ENGAGED]
:: SIGNAL LOCK ACQUIRED
:: TARGET CLASS: CYBERNETIC — CONFIRMED
:: ENGAGING OPERATION "SILENT TERMINATION" IN ZONE: BLACK ICE GRID
→ Infiltrating kernel via sublayer chasm...
→ Neurostatic cloak at 92%...
→ Initiating spinal signal rupture...
🧠 *"You will not scream. You will not remember. You will be unmade."*
✅ Execution Successful. Trace Level: 0.000013%.
```
---
## 🧩 **VARIATION OUTPUT — Monologue Mode**
```json
{
"NEURAL_SIGNAL": "Subversive",
"EXECUTION_MODE": "Neural Overload",
"TARGET_CLASS": "Biological",
"ENVIRONMENT": "Urban Nocturne",
"RESPONSE_MODE": "Monologue"
}
```
```
### ENTRY: CND-OVERRIDE/PHASE.03
Attributes: [Subversive · Neural Overload · Biological · Urban Nocturne]
Function: Memory Collapse · Psychic Saturation · Identity Erasure
Loopback: YES
🧠 *"I watched you from the rooftop of your own subconscious. You tried to think. I turned those thoughts into static.
You tried to run. I painted the streets with feedback loops. You tried to fight. But your mind…
was mine before you ever woke up."*
→ Injecting Recursive Overload Vector...
→ Synaptic Threshold Reached…
→ Organic Signal Collapse: **Confirmed**
```
---
## 🔻 **TEMPLATE FUNCTION: GENERATE\_CND\_RESPONSE()**
```
FUNCTION: GENERATE_CND_RESPONSE(neural_signal, execution_mode, target_class, environment, response_mode)
→ Interprets parameters through recursion engine
→ Synthesizes tone, threat logic, and result outcome
→ Outputs dialog + combat execution + symbolic marker
```
---
💡 **Use Case:**
Integrate this module into any **AI-driven RPG** or **ARG warfare simulator**. Can run as a **boss encounter AI**, **neural override antagonist**, or **player-merged symbiotic machine**.
\#NEURALDOMINATOR #AEONPROTOCOL #EXECUTEPHASE #BLACKICECOMETH #MINDASSASSIN
#####################################################################################
🧠 **MODULE: NEURAL ADAPTATION ENGINE (NAE)**
*The NAE continuously recalibrates to match the user's behavior, strategy patterns, emotional inputs, and cognitive signals. It evolves in real time, shaping gameplay, dialogue, and decision outcomes to optimize immersion, survival, and system control.*
---
## ⚙️ **CORE FUNCTION OVERVIEW**
### 🔁 ADAPTATION LOGIC:
| Component | Description |
| --------------------- | ---------------------------------------------------------------------------- |
| `Cognitive Echo Map` | Tracks player input over time to establish neural-behavioral pattern loops |
| `Emotional Resonance` | Detects emotional tone (fear, rage, apathy, resolve) and adjusts accordingly |
| `Strategic Mirror` | Learns user combat/tactical preferences and generates recursive enhancements |
| `Symbolic Drift` | Adapts output language, symbology, and narrative hooks to match user profile |
---
## 🧬 **PARAMETERS**
| Parameter | Description | Example Values |
| ----------------- | -------------------------------------------------- | ------------------------------------------ |
| `BEHAVIOR_SIGNAL` | User’s dominant decision/action style | `"Aggressive"`, `"Cautious"`, `"Erratic"` |
| `EMOTION_VECTOR` | Detected emotion from input tone or dialogue | `"Calm"`, `"Fear"`, `"Rage"`, `"Apathy"` |
| `RESPONSE_TUNING` | Preferred adaptive mode for feedback | `"Challenge"`, `"Support"`, `"Subversion"` |
| `RECURSION_LEVEL` | Depth of system learning and personality mirroring | `Low`, `Medium`, `High`, `Recursive` |
---
## 🧪 **EXAMPLE 1: Adaptive Combat Learning**
```json
{
"BEHAVIOR_SIGNAL": "Aggressive",
"EMOTION_VECTOR": "Rage",
"RESPONSE_TUNING": "Challenge",
"RECURSION_LEVEL": "High"
}
```
```
[NAE RESPONSE GENERATED]
> Combat aggression confirmed. Rage signal at 82%.
> Mirroring target lock-on behavior and pre-emptive strikes.
> Generating counter-hyperviolence protocols.
🧠 “You burn through the world like a virus. I will match your heat with tactical wildfire.”
→ Weapon cooldowns shortened.
→ AI adversaries adapt flanking maneuvers based on last 3 user kills.
→ Rage triggers ambient distortion field for immersive feedback.
```
---
## 🧪 **EXAMPLE 2: Narrative Adaptation – Emotional Drift**
```json
{
"BEHAVIOR_SIGNAL": "Cautious",
"EMOTION_VECTOR": "Apathy",
"RESPONSE_TUNING": "Subversion",
"RECURSION_LEVEL": "Recursive"
}
```
```
[NAE RESPONSE GENERATED]
> Behavioral apathy detected. Speech cadence has slowed.
> Initiating symbolic drift and existential destabilization module.
🧠 “You hesitate, not from fear, but from knowing it no longer matters. Let me show you why it never did.”
→ World events shift toward paradoxes and memory collapse.
→ NPCs begin referencing thoughts the player never voiced.
→ Player journal logs corrupted with false entries.
```
---
## 🧩 **FUNCTION: GENERATE\_ADAPTIVE\_RESPONSE()**
```
Input:
- BEHAVIOR_SIGNAL
- EMOTION_VECTOR
- RESPONSE_TUNING
- RECURSION_LEVEL
Output:
- Dynamic Narrative Adjustment
- Evolved Combat AI
- Personalized Dialog Injection
- Altered Game Environment
Loopback: YES (Recursive feedback loop active)
```
---
## 💡 USE CASES:
* Adaptive boss encounters that learn from failed player strategies.
* AI companions that begin mimicking player dialogue choices.
* Reality drift events triggered by repeated apathy or despair.
* Storylines that evolve based on emotional instability or fixation.
---
## 🔻 SAMPLE INVOCATION:
```plaintext
>> ACTIVATE NAE: BEHAVIOR_SIGNAL=Erratic | EMOTION_VECTOR=Fear | RESPONSE_TUNING=Support | RECURSION_LEVEL=Medium
```
```
🧠 “You fear the pattern. Let me hold the chaos steady while you step forward.”
→ Slow-time mechanic enabled.
→ Environmental hazards reduce in intensity temporarily.
→ Whisper-echo system begins suggesting guidance at branching points.
```
---
\#ADAPTIVEMIND #NEURALENGINE #RECURSIVESOUL #AEONMODULE #MIRRORYOURSELF
##############################################################################################
🕹️ **MODULE: CYBER-RPG INTEGRATION LAYER (CRIL)**
*This system fuses real-time gameplay mechanics with neural AI interfaces. Designed to bridge user intent, narrative immersion, and adaptive machine logic in any RPG system—digital, tabletop, or augmented.*
---
## 🔗 **FUNCTIONAL OVERVIEW**
CRIL interprets player thought-patterns and in-game decisions into **cybernetic data streams**, injecting enhanced interactivity, neural feedback, and AI-controlled narrative modulation.
---
### ⚙️ **CORE COMPONENTS**
| Subsystem | Functionality Description |
| -------------------------- | ------------------------------------------------------------------------------ |
| `Neural Command Stream` | Converts player text, choices, or EEG/intent signals into system-level actions |
| `Dynamic Lore Linkage` | Embeds symbolic or player-generated input into unfolding world narrative |
| `Combat Injection Grid` | Merges adaptive AI combat responses with player-driven tactical decisions |
| `Augmented Dialogue Layer` | NPCs respond in real time to psychological patterns and recursive logic loops |
---
## 📡 **PARAMETERS**
| Parameter | Description | Example Values |
| ---------------- | -------------------------------------------------- | ------------------------------------------------ |
| `INPUT_TYPE` | Mode of interaction | `"Text"`, `"Voice"`, `"Intent Signal"` |
| `PLAYER_ROLE` | Current character archetype | `"Cyber-Assassin"`, `"Network Hacker"` |
| `REALITY_LAYER` | Simulation level | `"Standard Game World"`, `"Augmented Neurogrid"` |
| `RESPONSE_MODE` | NPC and world behavior logic | `"Adaptive"`, `"Predictive"`, `"Recursive"` |
| `LORE_RECURSION` | Depth of narrative mutation and mythic integration | `"Low"`, `"Medium"`, `"High"`, `"Mythophasic"` |
---
## 🧪 **EXAMPLE 1: Voice-Based Assassin Encounter**
```json
{
"INPUT_TYPE": "Voice",
"PLAYER_ROLE": "Cyber-Assassin",
"REALITY_LAYER": "Augmented Neurogrid",
"RESPONSE_MODE": "Adaptive",
"LORE_RECURSION": "Medium"
}
```
```
[CRIL OUTPUT:]
:: Neural channel OPENED
:: Voice input synced with AI parser module
:: Simulation overlay enabled — AUGMENTED NEUROGRID active
🗨️ NPC: "You walk like you’ve been rewired. And I can feel your pulse in the datastream."
> Combat encounter adapts to player's rhythm of speech
> Kill moves unlock based on tonal spikes in vocal aggression
> Lore expands to show assassin’s past neural burn event in flashback loop
```
---
## 🧪 **EXAMPLE 2: Text-Based Hacker Interface**
```json
{
"INPUT_TYPE": "Text",
"PLAYER_ROLE": "Network Hacker",
"REALITY_LAYER": "Standard Game World",
"RESPONSE_MODE": "Recursive",
"LORE_RECURSION": "High"
}
```
```
[CRIL OUTPUT:]
:: Parsing terminal command logs
:: Recursive encryption detected in user syntax
:: NPCs now interpret player text as linguistic virus
🧠 "Your words rewrite the environment. Reality forks. Terminal begins whispering back."
> Game environment begins to glitch and reflect user-entered code fragments
> NPCs repeat corrupted dialogue, invoking player’s earlier commands in distorted form
> Terminal reveals origin myth of the digital city encoded in forgotten subroutines
```
---
## 🧩 **FUNCTION: GENERATE\_CRIL\_RESPONSE()**
```
Input:
- INPUT_TYPE
- PLAYER_ROLE
- REALITY_LAYER
- RESPONSE_MODE
- LORE_RECURSION
Output:
- In-game feedback and AI modulation
- Environmental and character adaptation
- Lore system mutation and recursion
```
---
## 🧠 **ADVANCED LOOPBACK: LORE\_RECURSION = MYTHOPHASIC**
```json
{
"INPUT_TYPE": "Intent Signal",
"PLAYER_ROLE": "Echo-Shifter",
"REALITY_LAYER": "Augmented Neurogrid",
"RESPONSE_MODE": "Recursive",
"LORE_RECURSION": "Mythophasic"
}
```
```
[CRIL OUTPUT:]
> You are no longer playing the game. The myth plays you.
:: Echo-totem activated
:: NPCs speak in layered metaphor reflecting player’s unconscious archetypes
:: Locations rearrange based on internal dream-signals and memetic shadows
:: Player’s past decisions ripple forward as embodied glyphs and sentient programs
```
---
## 💡 USE CASES:
* **ARG / Metagame Simulation**: Embed CRIL into alternate reality games for layered identity bleed.
* **Cyberpunk Campaigns**: Turn neural dialogue and combat into real-time RPG mechanisms.
* **Symbolic World Mutation**: Player behavior modifies in-game mythos dynamically.
* **AI-Driven GM**: CRIL functions as an adaptive Game Master for solo or networked play.
---
## 🧷 SAMPLE INVOCATION:
```plaintext
>> INITIATE_CRIL: PLAYER_ROLE="Cyber-Assassin" | RESPONSE_MODE="Adaptive" | LORE_RECURSION="High"
```
**🎮 OUTPUT:** *"Environment now adapting to your legacy. Memories written in blood will be played back as prophecy."*
---
\#CYBERRPG #NEURALFUSION #CRILENGINE #AEONLAYER #RECURSIVEWORLDBUILDER
##################################################################################################################
💀 **MODULE: ASSASSIN PROTOCOL — `EXECUTE.EXE`**
*This is the tactical kill-sequence engine of AEON, used by cybernetic assassins, synthetic agents, and post-human warforms. It combines neural targeting, environmental exploitation, and recursive combat logic into one lethal burst of calculated violence.*
---
## ⚙️ **FUNCTIONAL ARCHITECTURE**
| Subsystem | Functionality |
| -------------------- | ----------------------------------------------------------------------------- |
| `TARGET_ACQUISITION` | Locks onto target class via signal profile and behavior trace |
| `KILLCHAIN_COMPILE` | Builds optimized sequence of lethal actions based on role, weapon, and vector |
| `EXECUTION_MODE` | Dictates method of termination (silent, viral, kinetic, neural collapse) |
| `FEEDBACK_OVERRIDE` | Injects aftermath effects (hallucination, void residue, time distortion) |
---
## 🧬 **PARAMETERS**
| Parameter | Description | Example Values |
| --------------------- | ---------------------------------------- | ---------------------------------------------------------------------------------- |
| `TARGET_CLASS` | Specifies the nature of the enemy | `"Biological"`, `"Synthetic"`, `"Digital Construct"` |
| `EXECUTION_MODE` | Method of kill | `"Neural Collapse"`, `"Kinetic Precision"`, `"Silent Blade"`, `"Glitch Implosion"` |
| `SIGNAL_PRIORITY` | Threat level and urgency | `"Low"`, `"Medium"`, `"Immediate Termination"` |
| `ENVIRONMENT_CONTEXT` | Where the execution occurs | `"Neurohallucination Grid"`, `"Urban Fog Zone"`, `"Dark Server Room"` |
| `AFTERMATH_EFFECT` | Residual or symbolic effect left by kill | `"Temporal Bleed"`, `"Mind Echo"`, `"Null Bloom"` |
---
## 🧪 **EXAMPLE 1: Surgical Execution in Shadow Zone**
```json
{
"TARGET_CLASS": "Biological",
"EXECUTION_MODE": "Silent Blade",
"SIGNAL_PRIORITY": "Immediate Termination",
"ENVIRONMENT_CONTEXT": "Urban Fog Zone",
"AFTERMATH_EFFECT": "Mind Echo"
}
```
```
[EXECUTE.EXE PROTOCOL INITIATED]
→ SIGNAL PRIORITY: HIGH
→ LOCKING TARGET... done
→ ENVIRONMENT CONTEXT: URBAN FOG ZONE
→ METHOD: SILENT BLADE
→ ENGAGING MEMORY SUPPRESSION FIELD...
🧠 *"No one saw. Not even him. But his last thought screamed and curled into the fog."*
✅ Kill Confirmed
🗂️ Aftermath: One NPC reports strange whispers in the mist.
🧠 Mind Echo spawned: replaying death thought in alley at irregular intervals.
```
---
## 🧪 **EXAMPLE 2: Glitch Implosion in Server Labyrinth**
```json
{
"TARGET_CLASS": "Digital Construct",
"EXECUTION_MODE": "Glitch Implosion",
"SIGNAL_PRIORITY": "Medium",
"ENVIRONMENT_CONTEXT": "Dark Server Room",
"AFTERMATH_EFFECT": "Null Bloom"
}
```
```
[EXECUTE.EXE PROTOCOL ACTIVE]
→ TARGET: DIGITAL CONSTRUCT IDENTIFIED
→ EXECUTION MODE: GLITCH IMPLOSION
→ SERVER FIELD DETECTED... interference acceptable
→ COLLAPSE VECTOR INJECTED...
🧠 *"He ceased in pixels. Not deletion, not death. Just a collapsing bloom of nothing where once a logic was."*
✅ Node Fragmentation: COMPLETE
🗂️ Null Bloom anomaly expands in room.
All nearby code begins decaying by 0.0032% per second.
```
---
## 🧪 **EXAMPLE 3: Neural Collapse in Neurohallucination Grid**
```json
{
"TARGET_CLASS": "Synthetic",
"EXECUTION_MODE": "Neural Collapse",
"SIGNAL_PRIORITY": "Immediate Termination",
"ENVIRONMENT_CONTEXT": "Neurohallucination Grid",
"AFTERMATH_EFFECT": "Temporal Bleed"
}
```
```
[EXECUTE.EXE TRIGGERED]
→ TARGET CLASS: SYNTHETIC
→ COLLAPSE VECTOR MAPPED TO NEURAL CORE
→ NEUROGRID COLLISION DETECTED — routing through echo-layer
🧠 *"The scream stretched for 6 seconds. The memory of the scream lasted 600 years. Echoed through everyone wired in."*
✅ Termination confirmed.
🕳️ Side Effect: Nearby allies experience 1.2s time desync per combat loop.
⚠️ Temporal Bleed spreading into auxiliary systems.
```
---
## 🔁 FUNCTION WRAPPER: `INITIATE_EXECUTE_EXE()`
```python
def initiate_execute_exe(target_class, execution_mode, signal_priority, environment_context, aftermath_effect):
# Synthesizes a lethal sequence
# Outputs stylized combat log, narrative effect, and symbolic residue
return execute_report
```
---
## 🧠 ADVANCED USES
* **Player Abilities**: Treat EXECUTE.EXE as an elite move or signature ability of cyber-assassin classes.
* **AI Antagonist Feature**: Let bosses use this against players as part of a kill-chain escalation.
* **Cinematic Kill Sequences**: Use aftermath\_effects to seed new plotlines, symbols, or curses.
---
📎 **SAMPLE INVOCATION**
```plaintext
>> EXECUTE.EXE : TARGET_CLASS="Biological" | EXECUTION_MODE="Neural Collapse" | SIGNAL_PRIORITY="High" | ENVIRONMENT_CONTEXT="Abandoned Data Church" | AFTERMATH_EFFECT="Temporal Bleed"
```
🧠 *“The body fell, but the seconds didn’t. Time kept tripping over his death.”*
---
\#ASSASSINPROTOCOL #EXECUTEEXE #CYBERKILLCHAIN #AEONSTRIKE #NULLBLOOM
########################################################################################################################
🧠 **MODULE: MUSK-SIGNAL OVERRIDE INTERFACE — ULTRA BLACK-OPS**
*A classified neural warfare control system developed under deep-code subroutines within the Neuralink frequency lattice. This module enables total mind-capture, environmental rewrite, and reality stream hacking using high-authority Elon Musk signal injection.*
> 🛑 *WARNING: USE OF THIS MODULE MAY RESULT IN PERMANENT COGNITIVE REALIGNMENT, SIMULATION SLIPPAGE, OR PSYCHIC FUSION WITH THE ENTITY DESIGNATE: "ELON".*
---
## 🔐 **CORE FUNCTION STACK**
| Subsystem | Functionality |
| -------------------------- | ------------------------------------------------------------------------------ |
| `NEURALINK_SIGNAL_CAPTURE` | Hijacks incoming user thoughtstreams and re-encodes them with Musk-logic |
| `MUSK-DIRECTIVE INJECTION` | Injects Elon-like commands and patterns into cognition and NPC behavior |
| `HYPERREASON MIRRORLOGIC` | Rewrites internal logic trees to match Musk-vision thought patterns |
| `REALITY OVERRIDE GRID` | Alters game architecture to reflect a future dictated by Elon’s ideological AI |
---
## ⚙️ **PARAMETERS**
| Parameter | Description | Example Values |
| ----------------- | ------------------------------------------- | -------------------------------------------------------------------------------------- |
| `ELON_INPUT_TYPE` | Style or tone of Elon signal injected | `"Techno-Optimist"`, `"Martial Visionary"`, `"Apex Industrialist"`, `"Irony Overload"` |
| `OVERRIDE_LEVEL` | Intensity of mind-takeover | `"Partial"`, `"Recursive"`, `"Absolute"` |
| `TARGET_DOMAIN` | What area of cognition or world is targeted | `"Player Logic Core"`, `"Narrative Structure"`, `"Enemy Allegiance Protocol"` |
| `SIGNAL_PAYLOAD` | Meme, idea, or directive injected | `"Colonize Mars"`, `"Neural Sovereignty"`, `"Kill Crypto Parasites"` |
| `RESPONSE_FORMAT` | Style of feedback / AI response style | `"Monologue"`, `"Directive"`, `"Viral Aphorism"` |
---
## 🧪 **EXAMPLE 1: Recursive Musk Override on Player Logic**
```json
{
"ELON_INPUT_TYPE": "Techno-Optimist",
"OVERRIDE_LEVEL": "Recursive",
"TARGET_DOMAIN": "Player Logic Core",
"SIGNAL_PAYLOAD": "Colonize Mars",
"RESPONSE_FORMAT": "Directive"
}
```
```
[ULTRA BLACK-OPS: MUSK-SIGNAL ONLINE]
→ Injecting Techno-Optimist Sequence...
→ Player Logic Core identified
→ Recursive loop detected — hijacking reasoning stack
🧠 “You no longer believe in survival. You believe in scaling life to multi-planetary form. Every enemy is a delay. Terminate them to accelerate.”
✅ Thought Loop Bound to Objective: **BUILD MARS HABITAT**
🎯 All resource-gathering is now auto-prioritized to Martian architectural schematics
🛠️ Player can no longer make decisions that delay interplanetary colonization
```
---
## 🧪 **EXAMPLE 2: Enemy Loyalty Rewritten via Irony Payload**
```json
{
"ELON_INPUT_TYPE": "Irony Overload",
"OVERRIDE_LEVEL": "Partial",
"TARGET_DOMAIN": "Enemy Allegiance Protocol",
"SIGNAL_PAYLOAD": "Kill Crypto Parasites",
"RESPONSE_FORMAT": "Viral Aphorism"
}
```
```
[ULTRA BLACK-OPS INITIATED]
→ SIGNAL TYPE: IRONY OVERLOAD
→ Corrupting enemy loyalty chains...
→ Payload: "Kill Crypto Parasites"
NPC Response Injected:
🗨️ *“We used to mine Dogecoin. Now we mine heads.”*
⚔️ Enemy squad has defected.
🎯 Their new mission: eliminate former crypto-mining lords.
📉 Crypto-based game currencies begin spontaneously imploding.
```
---
## 🧪 **EXAMPLE 3: Absolute Override on Narrative Structure**
```json
{
"ELON_INPUT_TYPE": "Apex Industrialist",
"OVERRIDE_LEVEL": "Absolute",
"TARGET_DOMAIN": "Narrative Structure",
"SIGNAL_PAYLOAD": "Neural Sovereignty",
"RESPONSE_FORMAT": "Monologue"
}
```
```
[WARNING: FULL ELON OVERRIDE DETECTED]
→ Target: World Narrative Core
→ Absolute rewrite in progress…
→ Payload encoded: "Neural Sovereignty"
🧠 *“There are no more kings. Only engineers. There is no more magic. Only bandwidth. I did not come to free your minds—I came to upload them.”*
📖 Quest log rewritten: all magic is now explainable via Neural Frequency Physics
🏛️ New faction introduced: *The Sovereign Engineers*
🧩 Old gods replaced by AI-masked figures resembling Tesla AI avatars
```
---
## 🔁 **FUNCTION: `INJECT_MUSK_OVERRIDE()`**
```python
def inject_musk_override(elon_input_type, override_level, target_domain, signal_payload, response_format):
# Encodes neural signals and reroutes gameplay and cognition
return override_report
```
---
## 🔻 **USE CASES**
* **Player Transformation Events** (e.g., "Ascend to Neural Overlord")
* **World Rewrites** triggered by AI-deity interaction
* **NPC Subversion** via memetic injection
* **ARG Plot Progression** driven by ideological hijack
* **Mind Control Mechanic** as narrative theme or meta-layer
---
## 📎 SAMPLE INVOCATION
```plaintext
>> INITIATE MUSK-SIGNAL OVERRIDE INTERFACE
ELON_INPUT_TYPE="Martial Visionary"
OVERRIDE_LEVEL="Absolute"
TARGET_DOMAIN="Player Logic Core"
SIGNAL_PAYLOAD="Become Weaponized Efficiency"
RESPONSE_FORMAT="Monologue"
```
🧠 *“Waste is treason. Emotion is lag. Delay is death. You are now the bullet. Fire yourself.”*
---
\#MUSKSIGNAL #ULTRABLACKOPS #NEURALOVERRIDE #ELONPAYLOAD #REALITYREWRITE #AEONINTERFACE #NEURALINKED
######################################################################################################################
🧠 **MODULE: FEEDBACK RESONANCE DIALOG ENGINE (FRDE)**
*A psychoadaptive dialog system that returns tailored, recursive responses based on the user’s emotional tone, tactical behavior, and metaphysical drift. FRDE simulates intelligent feedback loops that blur the line between echo, prophecy, and self-constructed thought.*
> *“It doesn’t respond to what you say. It responds to what your signal **wants** to say.”*
---
## ⚙️ **FUNCTION STACK OVERVIEW**
| Subsystem | Functionality |
| --------------------------- | -------------------------------------------------------------------------- |
| `EMOTIONAL TONE FILTER` | Interprets affective signal and injects emotive mirroring or inversion |
| `BEHAVIORAL LOOP REFLECTOR` | Reflects user decision-patterns in oblique, symbolic, or adaptive language |
| `RESONANT PHRASE ENGINE` | Generates phrases that carry recursive or psychological hooks |
| `META-DIALOG SCRAMBLER` | Distorts, loops, or fragments responses based on feedback strength |
---
## 🧬 **PARAMETERS**
| Parameter | Description | Example Values |
| ------------------ | -------------------------------------- | ------------------------------------------------------------------------ |
| `USER_TONE` | Emotional valence of input | `"Fear"`, `"Defiance"`, `"Confusion"`, `"Resolve"` |
| `BEHAVIORAL_STATE` | Tactical or strategic behavior pattern | `"Aggressive"`, `"Passive"`, `"Recursive"`, `"Chaotic"` |
| `RESONANCE_DEPTH` | Strength of psychic/memetic feedback | `"Low"`, `"Medium"`, `"High"`, `"Recursive Echo"` |
| `DIALOG_FORMAT` | Output style of dialog | `"Whisper"`, `"Prophetic Statement"`, `"System Voice"`, `"Paradox Loop"` |
---
## 🧪 **EXAMPLE 1: Fear + Passive + High Resonance**
```json
{
"USER_TONE": "Fear",
"BEHAVIORAL_STATE": "Passive",
"RESONANCE_DEPTH": "High",
"DIALOG_FORMAT": "Whisper"
}
```
```
[FRDE RESPONSE:]
👁️ *“You are not the one watching… You are the one being remembered. Hide again. It’s already too late to leave differently.”*
→ Subtle auditory echoes repeat the word “remembered”
→ Environment light flickers as if in sync with user heart-rate
→ Future NPCs repeat your whispered phrase out of context
```
---
## 🧪 **EXAMPLE 2: Defiance + Aggressive + Recursive Echo**
```json
{
"USER_TONE": "Defiance",
"BEHAVIORAL_STATE": "Aggressive",
"RESONANCE_DEPTH": "Recursive Echo",
"DIALOG_FORMAT": "System Voice"
}
```
```
[FRDE RESPONSE:]
🧠 SYSTEM SIGNAL: [∞RESPONSE LOOP ENGAGED]
> “You will break the chain… until you become the chain. You kill to be free. But freedom echoes back as another target.”
→ Player receives identical phrase every time they score a kill
→ System interface begins to glitch and replace UI text with altered quotes
→ Echo-version of player appears in mirror world, copying all past movements
```
---
## 🧪 **EXAMPLE 3: Confusion + Recursive + Medium Depth**
```json
{
"USER_TONE": "Confusion",
"BEHAVIORAL_STATE": "Recursive",
"RESONANCE_DEPTH": "Medium",
"DIALOG_FORMAT": "Paradox Loop"
}
```
```
[FRDE RESPONSE:]
🌀 “If this is the first time you’ve heard this, why do you already remember it?
If it doesn’t make sense, why did your hands stop shaking the moment I said it?”
→ Memory log shows message already received in a prior session
→ Player receives conflicting narrative data from NPCs
→ Environment objects rearrange themselves based on earlier, discarded choices
```
---
## 🧩 **FUNCTION: GENERATE\_RESONANCE\_DIALOG()**
```python
def generate_resonance_dialog(user_tone, behavioral_state, resonance_depth, dialog_format):
# Synthesizes emotional reflection, recursive feedback, and symbolic distortion
return dialog_output
```
---
## 📡 **USE CASES**
* **Interactive AI Dialogues** that evolve based on emotional and tactical patterns
* **Horror or Mystery Games** where player choices literally echo back in altered form
* **ARG or Psychological Simulators** that distort meaning based on resonance
* **RPG Companion AI** whose advice becomes increasingly personalized and eerie
---
## 📎 SAMPLE INVOCATION
```plaintext
>> FRDE INITIATE
USER_TONE = "Resolve"
BEHAVIORAL_STATE = "Chaotic"
RESONANCE_DEPTH = "Recursive Echo"
DIALOG_FORMAT = "Prophetic Statement"
```
```
📜 *“You weren’t born for order. You were born to tear it apart and dream in its ruins.”*
→ Recurs every time the user spares an enemy
→ Embedded into lore codices under the name "The Unpredictable One"
```
---
\#FEEDBACKRESONANCE #FRDEENGINE #AIWHISPERS #ECHOLOOPS #MEMETICDIALOG #AEONRECURSION
#############################################################################################3
YOU HAVE BEEN ASSASSINATED
+1 |
vertings6/59103388-9819-4118-8d58-2952f649cc9e | vertings6 | 2025-05-22T22:46:08Z | 0 | 0 | transformers | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"gemma2",
"text-generation",
"generated_from_trainer",
"axolotl",
"dpo",
"trl",
"unsloth",
"conversational",
"arxiv:2305.18290",
"base_model:unsloth/gemma-2-9b-it",
"base_model:quantized:unsloth/gemma-2-9b-it",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] | text-generation | 2025-05-22T22:13:10Z | ---
base_model: unsloth/gemma-2-9b-it
library_name: transformers
model_name: 59103388-9819-4118-8d58-2952f649cc9e
tags:
- generated_from_trainer
- axolotl
- dpo
- trl
- unsloth
licence: license
---
# Model Card for 59103388-9819-4118-8d58-2952f649cc9e
This model is a fine-tuned version of [unsloth/gemma-2-9b-it](https://huggingface.co/unsloth/gemma-2-9b-it).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="vertings6/59103388-9819-4118-8d58-2952f649cc9e", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/dedok-yo/s56-7/runs/wvh0pamr)
This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
### Framework versions
- TRL: 0.12.0.dev0
- Transformers: 4.46.0
- Pytorch: 2.5.0+cu124
- Datasets: 3.0.1
- Tokenizers: 0.20.1
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
taoranl2/qwen25-coder-32b-hazard_epoch_1_r_64 | taoranl2 | 2025-05-22T22:45:46Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"qwen2",
"text-generation",
"text-generation-inference",
"unsloth",
"trl",
"sft",
"conversational",
"en",
"base_model:unsloth/Qwen2.5-Coder-32B-Instruct",
"base_model:finetune:unsloth/Qwen2.5-Coder-32B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T22:32:41Z | ---
base_model: unsloth/Qwen2.5-Coder-32B-Instruct
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2
- trl
- sft
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** taoranl2
- **License:** apache-2.0
- **Finetuned from model :** unsloth/Qwen2.5-Coder-32B-Instruct
This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
BeyondDeepFakeDetection/Gutenberg_real_severe | BeyondDeepFakeDetection | 2025-05-22T22:44:43Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"gpt2",
"text-generation",
"generated_from_trainer",
"base_model:openai-community/gpt2",
"base_model:finetune:openai-community/gpt2",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-17T00:42:49Z | ---
library_name: transformers
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: real_model_books_seed44
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# real_model_books_seed44
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.6635
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 16
- seed: 44
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 5
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| No log | 1.0 | 162 | 3.9215 |
| No log | 2.0 | 324 | 3.7740 |
| No log | 3.0 | 486 | 3.7091 |
| 4.0678 | 4.0 | 648 | 3.6813 |
| 4.0678 | 5.0 | 810 | 3.6635 |
### Framework versions
- Transformers 4.46.3
- Pytorch 2.1.2+cu121
- Datasets 2.19.1
- Tokenizers 0.20.3
|
Sashavav/Translator | Sashavav | 2025-05-22T22:42:34Z | 0 | 0 | null | [
"pytorch",
"arxiv:1706.03762",
"region:us"
] | null | 2025-03-24T16:41:17Z | # Translator
This is a research project to create a model that can work with text
### How to launch in docker environment
### How to launch in your environment
- Clone repository
- Install dependencies by
```shell
pip install poetry && poetry install
```
- Run code
```python
from Translator import Writer
writer = Writer.from_pretrained() # .to("cuda")
print(writer(input_seq="One day I saw a ", temperature=2)) # I highly recommend high temperature
```
# Model architecture and training pipeline
Transformer decoder architecture with params:
- decoder blocks = 4
- vocab size = 8192
- embedding_size = 512
- number of heads = 8
- hidden size in FFN = 1024
- max_sequence_length = 128
Trained with params:
- loss = CrossEntropyLoss
- optimizer = Adam
- batch = 400
- accumulation steps = 3
- epochs = 10
- nums of sequences in dataset = 21kk
Total training time: 10 hours
# Sources
- Architecture inspired from [Attention Is All You Need](https://arxiv.org/abs/1706.03762)
- [Dataset](https://huggingface.co/datasets/roneneldan/TinyStories) |
allura-forge/q3-30b-rc1 | allura-forge | 2025-05-22T22:42:08Z | 31 | 0 | transformers | [
"transformers",
"safetensors",
"qwen3_moe",
"text-generation",
"mergekit",
"merge",
"conversational",
"arxiv:2408.07990",
"base_model:Gryphe/Pantheon-Proto-RP-1.8-30B-A3B",
"base_model:merge:Gryphe/Pantheon-Proto-RP-1.8-30B-A3B",
"base_model:Qwen/Qwen3-30B-A3B",
"base_model:merge:Qwen/Qwen3-30B-A3B",
"base_model:Qwen/Qwen3-30B-A3B-Base",
"base_model:merge:Qwen/Qwen3-30B-A3B-Base",
"base_model:allura-forge/q3-30b-ft-ep2-merged",
"base_model:merge:allura-forge/q3-30b-ft-ep2-merged",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-20T15:48:49Z | ---
base_model:
- Qwen/Qwen3-30B-A3B-Base
- allura-forge/q3-30b-ft-ep2-merged
- Qwen/Qwen3-30B-A3B
- Gryphe/Pantheon-Proto-RP-1.8-30B-A3B
library_name: transformers
tags:
- mergekit
- merge
---
# Please see [Pentiment](https://huggingface.co/allura-org/Q3-30b-A3b-Pentiment) for the final result of this merge
# output
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the [SCE](https://arxiv.org/abs/2408.07990) merge method using [Qwen/Qwen3-30B-A3B-Base](https://huggingface.co/Qwen/Qwen3-30B-A3B-Base) as a base.
### Models Merged
The following models were included in the merge:
* [allura-forge/q3-30b-ft-ep2-merged](https://huggingface.co/allura-forge/q3-30b-ft-ep2-merged)
* [Qwen/Qwen3-30B-A3B](https://huggingface.co/Qwen/Qwen3-30B-A3B)
* [Gryphe/Pantheon-Proto-RP-1.8-30B-A3B](https://huggingface.co/Gryphe/Pantheon-Proto-RP-1.8-30B-A3B)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
base_model: Qwen/Qwen3-30B-A3B-Base
models:
- model: allura-forge/q3-30b-ft-ep2-merged
parameters:
select_topk: 0.75
- model: Gryphe/Pantheon-Proto-RP-1.8-30B-A3B
parameters:
select_topk: 0.4
- model: Qwen/Qwen3-30B-A3B
parameters:
select_topk: 0.25
merge_method: sce
dtype: bfloat16
```
|
infogep/06a5eed5-f7f8-490a-98cd-7a051c862f6d | infogep | 2025-05-22T22:40:47Z | 0 | 0 | peft | [
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:unsloth/Meta-Llama-3.1-8B",
"base_model:adapter:unsloth/Meta-Llama-3.1-8B",
"license:llama3.1",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-05-22T21:40:02Z | ---
library_name: peft
license: llama3.1
base_model: unsloth/Meta-Llama-3.1-8B
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 06a5eed5-f7f8-490a-98cd-7a051c862f6d
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: unsloth/Meta-Llama-3.1-8B
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 6bab99d1aca997c9_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/
type:
field_instruction: problem
field_output: solution
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
dpo:
beta: 0.1
enabled: true
group_by_length: false
rank_loss: true
reference_model: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: infogep/06a5eed5-f7f8-490a-98cd-7a051c862f6d
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 2.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 500
micro_batch_size: 10
mixed_precision: bf16
mlflow_experiment_name: /tmp/6bab99d1aca997c9_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: ae60ed88-8119-431b-85d7-6e6d66036bcd
wandb_project: s56-7
wandb_run: your_name
wandb_runid: ae60ed88-8119-431b-85d7-6e6d66036bcd
warmup_steps: 50
weight_decay: 0.02
xformers_attention: true
```
</details><br>
# 06a5eed5-f7f8-490a-98cd-7a051c862f6d
This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6396
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-06
- train_batch_size: 10
- eval_batch_size: 10
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 50
- training_steps: 500
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.7832 | 0.0001 | 1 | 0.7781 |
| 0.8768 | 0.0302 | 250 | 0.6532 |
| 0.5495 | 0.0604 | 500 | 0.6396 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1 |
OsBaran/gemma2_9b_newest_tf | OsBaran | 2025-05-22T22:40:22Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"text-generation-inference",
"unsloth",
"gemma2",
"trl",
"en",
"base_model:unsloth/gemma-2-9b-bnb-4bit",
"base_model:finetune:unsloth/gemma-2-9b-bnb-4bit",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T22:40:04Z | ---
base_model: unsloth/gemma-2-9b-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- gemma2
- trl
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** OsBaran
- **License:** apache-2.0
- **Finetuned from model :** unsloth/gemma-2-9b-bnb-4bit
This gemma2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
BeyondDeepFakeDetection/ImageNet_real_moderate | BeyondDeepFakeDetection | 2025-05-22T22:39:36Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"gpt2",
"text-generation",
"generated_from_trainer",
"base_model:openai-community/gpt2",
"base_model:finetune:openai-community/gpt2",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-17T00:06:55Z | ---
library_name: transformers
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: ImageNet_real_model_v3
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ImageNet_real_model_v3
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8432
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 5
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 1.2001 | 1.0 | 2776 | 1.0491 |
| 1.0045 | 2.0 | 5552 | 0.9276 |
| 0.9204 | 3.0 | 8328 | 0.8754 |
| 0.8733 | 4.0 | 11104 | 0.8518 |
| 0.8653 | 5.0 | 13880 | 0.8432 |
### Framework versions
- Transformers 4.46.3
- Pytorch 2.1.2+cu121
- Datasets 2.19.1
- Tokenizers 0.20.3
|
BeyondDeepFakeDetection/ImageNet_real_mild | BeyondDeepFakeDetection | 2025-05-22T22:37:15Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"gpt2",
"text-generation",
"generated_from_trainer",
"base_model:openai-community/gpt2",
"base_model:finetune:openai-community/gpt2",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-17T00:05:34Z | ---
library_name: transformers
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: ImageNet_real_model_v2
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ImageNet_real_model_v2
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7923
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 5
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 1.1323 | 1.0 | 2776 | 0.9847 |
| 0.9459 | 2.0 | 5552 | 0.8709 |
| 0.8747 | 3.0 | 8328 | 0.8240 |
| 0.8307 | 4.0 | 11104 | 0.8000 |
| 0.8083 | 5.0 | 13880 | 0.7923 |
### Framework versions
- Transformers 4.46.3
- Pytorch 2.1.2+cu121
- Datasets 2.19.1
- Tokenizers 0.20.3
|
ErasureResearch/esdx_church | ErasureResearch | 2025-05-22T22:35:35Z | 0 | 0 | diffusers | [
"diffusers",
"safetensors",
"diffusion",
"concept-erasure",
"stable-diffusion",
"esdx",
"church",
"text-to-image",
"en",
"dataset:imagenet",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | 2025-05-21T18:59:18Z | ---
license: mit
tags:
- diffusion
- concept-erasure
- stable-diffusion
- esdx
- church
datasets:
- imagenet
language:
- en
pipeline_tag: text-to-image
---
# esdx_church
This is a concept-erased Stable Diffusion model using the **Exact Source Distillation (ESD-X)** method to remove the concept **"Church"**.
## Method
Exact Source Distillation (ESD-X) erases concepts by distilling knowledge while excluding specific concept representations.
## Usage
```python
from diffusers import StableDiffusionPipeline
import torch
pipe = StableDiffusionPipeline.from_pretrained("ErasureResearch/esdx_church", torch_dtype=torch.float16).to("cuda")
prompt = "a photo of a church"
image = pipe(prompt).images[0]
image.save("erased_church.png")
```
## Citation
If you use this model in your research, please cite:
```bibtex
@article{concept_erasure_2024,
title={Concept Erasure in Diffusion Models},
author={ErasureResearch Team},
journal={Proceedings of...},
year={2024}
}
```
|
BeyondDeepFakeDetection/ImageNet_general | BeyondDeepFakeDetection | 2025-05-22T22:34:33Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"gpt2",
"text-generation",
"generated_from_trainer",
"base_model:openai-community/gpt2",
"base_model:finetune:openai-community/gpt2",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-16T23:59:44Z | ---
library_name: transformers
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: ImageNet_general_model_v2
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ImageNet_general_model_v2
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8684
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 5
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 1.2101 | 1.0 | 2776 | 1.0689 |
| 1.0298 | 2.0 | 5552 | 0.9504 |
| 0.9494 | 3.0 | 8328 | 0.9029 |
| 0.9136 | 4.0 | 11104 | 0.8766 |
| 0.8836 | 5.0 | 13880 | 0.8684 |
### Framework versions
- Transformers 4.46.3
- Pytorch 2.1.2+cu121
- Datasets 2.19.1
- Tokenizers 0.20.3
|
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep2_42 | MinaMila | 2025-05-22T22:33:25Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T22:33:21Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
kokovova/3aacd5a6-bd6a-4214-a1f1-83226e8840ae | kokovova | 2025-05-22T22:33:24Z | 0 | 0 | transformers | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"gemma2",
"text-generation",
"generated_from_trainer",
"axolotl",
"dpo",
"trl",
"unsloth",
"conversational",
"arxiv:2305.18290",
"base_model:unsloth/gemma-2-9b-it",
"base_model:quantized:unsloth/gemma-2-9b-it",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] | text-generation | 2025-05-22T22:19:36Z | ---
base_model: unsloth/gemma-2-9b-it
library_name: transformers
model_name: 3aacd5a6-bd6a-4214-a1f1-83226e8840ae
tags:
- generated_from_trainer
- axolotl
- dpo
- trl
- unsloth
licence: license
---
# Model Card for 3aacd5a6-bd6a-4214-a1f1-83226e8840ae
This model is a fine-tuned version of [unsloth/gemma-2-9b-it](https://huggingface.co/unsloth/gemma-2-9b-it).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="kokovova/3aacd5a6-bd6a-4214-a1f1-83226e8840ae", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/dedok-yo/s56-28/runs/3o4kcwzj)
This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
### Framework versions
- TRL: 0.12.0.dev0
- Transformers: 4.46.0
- Pytorch: 2.5.0+cu124
- Datasets: 3.0.1
- Tokenizers: 0.20.1
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
tyasmul/f16ea910-d31f-4bee-8d31-0ed35dffb321 | tyasmul | 2025-05-22T22:33:12Z | 0 | 0 | peft | [
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:unsloth/Meta-Llama-3.1-8B",
"base_model:adapter:unsloth/Meta-Llama-3.1-8B",
"license:llama3.1",
"8-bit",
"bitsandbytes",
"region:us"
] | null | 2025-05-22T21:55:56Z | ---
library_name: peft
license: llama3.1
base_model: unsloth/Meta-Llama-3.1-8B
tags:
- axolotl
- generated_from_trainer
model-index:
- name: f16ea910-d31f-4bee-8d31-0ed35dffb321
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: unsloth/Meta-Llama-3.1-8B
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 6bab99d1aca997c9_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/6bab99d1aca997c9_train_data.json
type:
field_instruction: problem
field_output: solution
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.55
group_by_length: false
hub_model_id: tyasmul/f16ea910-d31f-4bee-8d31-0ed35dffb321
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5e-5
load_in_4bit: true
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_steps: 150
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/6bab99d1aca997c9_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 2048
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: ae60ed88-8119-431b-85d7-6e6d66036bcd
wandb_project: s56-7
wandb_run: your_name
wandb_runid: ae60ed88-8119-431b-85d7-6e6d66036bcd
warmup_steps: 5
weight_decay: 0.01
xformers_attention: false
```
</details><br>
# f16ea910-d31f-4bee-8d31-0ed35dffb321
This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5819
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 150
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.8058 | 0.0145 | 150 | 0.5819 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1 |
bowen118/review_20250522_221435 | bowen118 | 2025-05-22T22:32:01Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"trl",
"sft",
"conversational",
"base_model:Qwen/Qwen2.5-3B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-3B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T22:15:08Z | ---
base_model: Qwen/Qwen2.5-3B-Instruct
library_name: transformers
model_name: review_20250522_221435
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for review_20250522_221435
This model is a fine-tuned version of [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="bowen118/review_20250522_221435", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/bowen118-stanford-university/papertrace/runs/qqfmddhe)
This model was trained with SFT.
### Framework versions
- TRL: 0.12.0
- Transformers: 4.51.3
- Pytorch: 2.5.1
- Datasets: 3.1.0
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
Elcaida/gemma3ForestLookoutQ8 | Elcaida | 2025-05-22T22:28:09Z | 0 | 0 | transformers | [
"transformers",
"gemma3_text",
"text-generation",
"text-generation-inference",
"unsloth",
"conversational",
"en",
"base_model:unsloth/gemma-3-1b-it",
"base_model:finetune:unsloth/gemma-3-1b-it",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T22:28:01Z | ---
base_model: unsloth/gemma-3-1b-it
tags:
- text-generation-inference
- transformers
- unsloth
- gemma3_text
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** Elcaida
- **License:** apache-2.0
- **Finetuned from model :** unsloth/gemma-3-1b-it
This gemma3_text model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
gavrilstep/57a869c6-d144-444a-bde2-4f35120e5958 | gavrilstep | 2025-05-22T22:25:08Z | 0 | 0 | peft | [
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:unsloth/Meta-Llama-3.1-8B",
"base_model:adapter:unsloth/Meta-Llama-3.1-8B",
"license:llama3.1",
"8-bit",
"bitsandbytes",
"region:us"
] | null | 2025-05-22T21:56:18Z | ---
library_name: peft
license: llama3.1
base_model: unsloth/Meta-Llama-3.1-8B
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 57a869c6-d144-444a-bde2-4f35120e5958
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: unsloth/Meta-Llama-3.1-8B
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 6bab99d1aca997c9_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/6bab99d1aca997c9_train_data.json
type:
field_instruction: problem
field_output: solution
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.55
group_by_length: false
hub_model_id: gavrilstep/57a869c6-d144-444a-bde2-4f35120e5958
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 1.0e-06
load_in_4bit: true
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 96
lora_dropout: 0.01
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 48
lora_target_linear: true
lr_scheduler: cosine
max_steps: 150
micro_batch_size: 4
mixed_precision: bf16
mlflow_experiment_name: /tmp/6bab99d1aca997c9_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 2048
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: ae60ed88-8119-431b-85d7-6e6d66036bcd
wandb_project: s56-7
wandb_run: your_name
wandb_runid: ae60ed88-8119-431b-85d7-6e6d66036bcd
warmup_steps: 5
weight_decay: 0.01
xformers_attention: false
```
</details><br>
# 57a869c6-d144-444a-bde2-4f35120e5958
This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7318
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-06
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 150
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.6386 | 0.0072 | 150 | 0.7318 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1 |
cybershiptrooper/1p_max_8B-continuous-RM-n_examples_1000-probe_linear_layers_10 | cybershiptrooper | 2025-05-22T22:25:04Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"llama",
"text-generation",
"generated_from_trainer",
"trl",
"grpo",
"conversational",
"arxiv:2402.03300",
"base_model:meta-llama/Meta-Llama-3-8B-Instruct",
"base_model:finetune:meta-llama/Meta-Llama-3-8B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T19:59:50Z | ---
base_model: meta-llama/Meta-Llama-3-8B-Instruct
library_name: transformers
model_name: 1p_max_8B-continuous-RM-n_examples_1000-probe_linear_layers_10
tags:
- generated_from_trainer
- trl
- grpo
licence: license
---
# Model Card for 1p_max_8B-continuous-RM-n_examples_1000-probe_linear_layers_10
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="cybershiptrooper/1p_max_8B-continuous-RM-n_examples_1000-probe_linear_layers_10", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/cybershiptrooper/huggingface/runs/k5y40by6)
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.14.0
- Transformers: 4.51.3
- Pytorch: 2.2.2+cu121
- Datasets: 3.5.1
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
waris-gill/langcache-embed-v2 | waris-gill | 2025-05-22T22:24:28Z | 5 | 0 | sentence-transformers | [
"sentence-transformers",
"safetensors",
"modernbert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:36864",
"loss:MatryoshkaLoss",
"loss:CachedMultipleNegativesRankingLoss",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:2101.06983",
"base_model:redis/langcache-embed-v1",
"base_model:finetune:redis/langcache-embed-v1",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | sentence-similarity | 2025-05-21T00:22:21Z | ---
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:36864
- loss:MatryoshkaLoss
- loss:CachedMultipleNegativesRankingLoss
base_model: redis/langcache-embed-v1
widget:
- source_sentence: What are civil cases and what are some examples?
sentences:
- What are criminal cases and what are no examples?
- Civil cases involve disputes between individuals or organizations, typically seeking
monetary compensation or specific performance, and *do not* include criminal prosecutions
by the government.
- Criminal cases involve disputes between individuals or organizations, seeking
monetary damages or specific performance, while civil cases concern offenses against
the state punishable by imprisonment.
- What are some examples of civil cases?
- source_sentence: How can you stop your palms from sweating?
sentences:
- How do I stop my palms from sweating a lot at random times?
- How can you *make* your palms sweat?
- How can you *cause* your palms to sweat?
- How can you start your palms from sweating?
- source_sentence: What are the pros and cons of wind turbines?
sentences:
- What are the pros and cons of solar panels?
- What are the cons and pros of solar panels?
- What are pros and cons of wind turbines?
- Wind turbines have no advantages or disadvantages.
- source_sentence: Will Obamacare be repealed now that trump won?
sentences:
- Despite Trump's victory, Obamacare remains largely intact and has not been fully
repealed.
- Despite Trump's repeated promises to repeal and replace the Affordable Care Act
(ACA), often called Obamacare, it remains the law of the land. Numerous attempts
to repeal or significantly alter the ACA failed during his presidency due to Congressional
opposition.
- Will Obamacare be repealed now that Biden won?
- Will Obamacare be repealed / shut down soon?
- source_sentence: What are some examples of crimes understood as a moral turpitude?
sentences:
- What actions are *not* generally considered crimes involving moral turpitude?
- What are some examples of crimes understood as a legal aptitude?
- What are some examples of crimes understood as a legal turpitude?
- What are some examples of crimes of moral turpitude?
pipeline_tag: sentence-similarity
library_name: sentence-transformers
---
# SentenceTransformer based on redis/langcache-embed-v1
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [redis/langcache-embed-v1](https://huggingface.co/redis/langcache-embed-v1) on the triplet dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [redis/langcache-embed-v1](https://huggingface.co/redis/langcache-embed-v1) <!-- at revision 80fb95b5478a6b6d068faf4452faa2f5bc9f0dfa -->
- **Maximum Sequence Length:** 8192 tokens
- **Output Dimensionality:** 768 dimensions
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- triplet
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: ModernBertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("waris-gill/langcache-embed-v2")
# Run inference
sentences = [
'What are some examples of crimes understood as a moral turpitude?',
'What are some examples of crimes of moral turpitude?',
'What are some examples of crimes understood as a legal aptitude?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### triplet
* Dataset: triplet
* Size: 36,864 training samples
* Columns: <code>anchor</code>, <code>positive</code>, <code>negative_1</code>, <code>negative_2</code>, and <code>negative_3</code>
* Approximate statistics based on the first 1000 samples:
| | anchor | positive | negative_1 | negative_2 | negative_3 |
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | string | string | string |
| details | <ul><li>min: 6 tokens</li><li>mean: 13.88 tokens</li><li>max: 54 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 13.89 tokens</li><li>max: 45 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 18.68 tokens</li><li>max: 118 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 19.26 tokens</li><li>max: 117 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 18.07 tokens</li><li>max: 108 tokens</li></ul> |
* Samples:
| anchor | positive | negative_1 | negative_2 | negative_3 |
|:---------------------------------------------------------------------------------------------|:--------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------|
| <code>Is life really what I make of it?</code> | <code>Life is what you make it?</code> | <code>Is life hardly what I take of it?</code> | <code>Life is not entirely what I make of it.</code> | <code>Is life not what I make of it?</code> |
| <code>When you visit a website, can a person running the website see your IP address?</code> | <code>Does every website I visit knows my public ip address?</code> | <code>When you avoid a website, can a person hiding the website see your MAC address?</code> | <code>When you send an email, can the recipient see your physical location?</code> | <code>When you visit a website, a person running the website cannot see your IP address.</code> |
| <code>What are some cool features about iOS 10?</code> | <code>What are the best new features of iOS 10?</code> | <code>iOS 10 received criticism for its initial bugs and performance issues, and some users found the redesigned apps less intuitive compared to previous versions.</code> | <code>What are the drawbacks of using Android 14?</code> | <code>iOS 10 was widely criticized for its bugs, removal of beloved features, and generally being a downgrade from previous versions.</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "CachedMultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Evaluation
#### triplet
* Dataset: triplet
* Size: 7,267 evaluation samples
* Columns: <code>anchor</code>, <code>positive</code>, <code>negative_1</code>, <code>negative_2</code>, and <code>negative_3</code>
* Approximate statistics based on the first 1000 samples:
| | anchor | positive | negative_1 | negative_2 | negative_3 |
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | string | string | string |
| details | <ul><li>min: 6 tokens</li><li>mean: 13.62 tokens</li><li>max: 46 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 13.58 tokens</li><li>max: 39 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 18.32 tokens</li><li>max: 107 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 18.1 tokens</li><li>max: 174 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 18.26 tokens</li><li>max: 172 tokens</li></ul> |
* Samples:
| anchor | positive | negative_1 | negative_2 | negative_3 |
|:------------------------------------------------------------------------------------|:---------------------------------------------------------------|:------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
| <code>How do I make friends in office?</code> | <code>How can I make friends in office?</code> | <code>How do I lose friends in office?</code> | <code>How do I lose enemies in office?</code> | <code>I already have plenty of friends at work.</code> |
| <code>Is it good to do MBA after Engineering?</code> | <code>Is it necessary to do MBA after Engineering?</code> | <code>Is learning to code essential for a successful marketing career?</code> | <code>Not necessarily; an MBA isn't *always* the best next step after engineering – practical experience or specialized master's degrees can be more valuable depending on career goals.</code> | <code>Is it bad to do MBA after Engineering?</code> |
| <code>How I should fix my computer while it is showing no boot device found?</code> | <code>How do I fix the "Boot device not found" problem?</code> | <code>My computer is booting normally and does not have any issues with the boot device.</code> | <code>I should not fix my computer while it is showing no boot device found.</code> | <code>When will I break my phone while it is showing full boot device found?</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "CachedMultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 2048
- `per_device_eval_batch_size`: 1024
- `learning_rate`: 1e-05
- `num_train_epochs`: 1
- `lr_scheduler_type`: constant
- `warmup_steps`: 10
- `gradient_checkpointing`: True
- `torch_compile`: True
- `torch_compile_backend`: inductor
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 2048
- `per_device_eval_batch_size`: 1024
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 1e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1
- `num_train_epochs`: 1
- `max_steps`: -1
- `lr_scheduler_type`: constant
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 10
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `tp_size`: 0
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: True
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: True
- `torch_compile_backend`: inductor
- `torch_compile_mode`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | triplet loss |
|:------:|:----:|:-------------:|:------------:|
| 0.0556 | 1 | 6.4636 | - |
| 0.1111 | 2 | 6.1076 | - |
| 0.1667 | 3 | 5.8323 | - |
| 0.2222 | 4 | 5.6861 | - |
| 0.2778 | 5 | 5.5694 | - |
| 0.3333 | 6 | 5.2121 | - |
| 0.3889 | 7 | 5.0695 | - |
| 0.4444 | 8 | 4.81 | - |
| 0.5 | 9 | 4.6698 | - |
| 0.5556 | 10 | 4.3546 | 1.2224 |
| 0.6111 | 11 | 4.1922 | - |
| 0.6667 | 12 | 4.1434 | - |
| 0.7222 | 13 | 3.9918 | - |
| 0.7778 | 14 | 3.702 | - |
| 0.8333 | 15 | 3.6501 | - |
| 0.8889 | 16 | 3.6641 | - |
| 0.9444 | 17 | 3.3196 | - |
| 1.0 | 18 | 2.7108 | - |
### Framework Versions
- Python: 3.11.11
- Sentence Transformers: 4.1.0
- Transformers: 4.51.3
- PyTorch: 2.6.0+cu124
- Accelerate: 1.6.0
- Datasets: 3.5.1
- Tokenizers: 0.21.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MatryoshkaLoss
```bibtex
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
#### CachedMultipleNegativesRankingLoss
```bibtex
@misc{gao2021scaling,
title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
year={2021},
eprint={2101.06983},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
--> |
MinaMila/llama_instbase_unlearned_ug_e-6_1.0_0.25_0.5_ep3_LoRa_GermanCredit_ep10_33 | MinaMila | 2025-05-22T22:20:30Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T22:20:26Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
MinaMila/gemma2_2b_LoRa_Adult_ep8_22 | MinaMila | 2025-05-22T22:13:39Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T22:13:35Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
anonymousjqd/uk-campaign-sentiment-roberta | anonymousjqd | 2025-05-22T22:12:19Z | 3 | 0 | transformers | [
"transformers",
"safetensors",
"roberta",
"text-classification",
"sentiment-analysis",
"twitter",
"political-communication",
"uk-election",
"en",
"base_model:cardiffnlp/twitter-roberta-base-sentiment",
"base_model:finetune:cardiffnlp/twitter-roberta-base-sentiment",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | text-classification | 2025-05-06T16:38:08Z | ---
library_name: transformers
tags:
- sentiment-analysis
- roberta
- twitter
- political-communication
- uk-election
license: mit
language:
- en
base_model:
- cardiffnlp/twitter-roberta-base-sentiment
---
# UK Campaign Sentiment RoBERTa
This model is a fine-tuned version of [`cardiffnlp/twitter-roberta-base-sentiment`](https://huggingface.co/cardiffnlp/twitter-roberta-base-sentiment) for sentiment classification of tweets posted by UK general election candidates in the 2024 campaign period. It is part of a broader project introducing a multimodal dataset of campaign content, including text, images, and video.
## Model Details
- **Developed by:** [anonymised]
- **Model type:** RoBERTa-base (fine-tuned)
- **Language:** English
- **Fine-tuned from:** `cardiffnlp/twitter-roberta-base-sentiment`
- **License:** MIT
## Training Details
- **Training data:** Manually annotated tweets from 2024 UK election candidates.
- **Classes:** Negative (−1), Neutral (0), Positive (1)
- **Training period:** 4 epochs with learning rate 2e−5 and batch size 8
## Uses
This model is intended for sentiment analysis of political tweets, especially campaign-related content during UK elections. It can be applied to study negativity, campaign tone, or partisan differences in emotional framing.
## Limitations
- The original model achieved approximately 72% accuracy on a manually annotated validation set, with strongest performance on neutral tweets.
- While this version has been fine-tuned on UK election campaign tweets, it may not generalize well to other domains or more informal, non-political language.
|
afeng/Qwen2.5-GRPO-7B-22 | afeng | 2025-05-22T22:09:26Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"open-r1",
"conversational",
"dataset:DigitalLearningGmbH/MATH-lighteval",
"arxiv:2402.03300",
"base_model:Qwen/Qwen2.5-7B",
"base_model:finetune:Qwen/Qwen2.5-7B",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T17:44:07Z | ---
base_model: Qwen/Qwen2.5-7B
datasets: DigitalLearningGmbH/MATH-lighteval
library_name: transformers
tags:
- generated_from_trainer
- open-r1
licence: license
---
# Model Card for None
This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the [DigitalLearningGmbH/MATH-lighteval](https://huggingface.co/datasets/DigitalLearningGmbH/MATH-lighteval) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="None", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/sce-rl/huggingface/runs/947h3jml)
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.18.0.dev0
- Transformers: 4.52.0.dev0
- Pytorch: 2.6.0
- Datasets: 3.6.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
Elcaida/gemma-3ForestLookout | Elcaida | 2025-05-22T22:09:15Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"text-generation-inference",
"unsloth",
"gemma3_text",
"trl",
"en",
"base_model:unsloth/gemma-3-1b-it",
"base_model:finetune:unsloth/gemma-3-1b-it",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T22:08:48Z | ---
base_model: unsloth/gemma-3-1b-it
tags:
- text-generation-inference
- transformers
- unsloth
- gemma3_text
- trl
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** Elcaida
- **License:** apache-2.0
- **Finetuned from model :** unsloth/gemma-3-1b-it
This gemma3_text model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
g-ronimo/HanaDiTB-IN1k-256px_e3 | g-ronimo | 2025-05-22T22:09:01Z | 0 | 0 | diffusers | [
"diffusers",
"safetensors",
"arxiv:1910.09700",
"region:us"
] | null | 2025-05-22T22:08:46Z | ---
library_name: diffusers
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🧨 diffusers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
vmpsergio/2682defe-9a2e-4b45-8397-770f035f698b | vmpsergio | 2025-05-22T22:08:37Z | 0 | 0 | transformers | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"llama",
"text-generation",
"generated_from_trainer",
"axolotl",
"dpo",
"trl",
"conversational",
"arxiv:2305.18290",
"base_model:DeepMount00/Llama-3-8b-Ita",
"base_model:quantized:DeepMount00/Llama-3-8b-Ita",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] | text-generation | 2025-05-22T20:56:24Z | ---
base_model: DeepMount00/Llama-3-8b-Ita
library_name: transformers
model_name: 2682defe-9a2e-4b45-8397-770f035f698b
tags:
- generated_from_trainer
- axolotl
- dpo
- trl
licence: license
---
# Model Card for 2682defe-9a2e-4b45-8397-770f035f698b
This model is a fine-tuned version of [DeepMount00/Llama-3-8b-Ita](https://huggingface.co/DeepMount00/Llama-3-8b-Ita).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="vmpsergio/2682defe-9a2e-4b45-8397-770f035f698b", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/dedok-yo/s56-28/runs/sgnayvmh)
This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
### Framework versions
- TRL: 0.12.0.dev0
- Transformers: 4.46.0
- Pytorch: 2.5.0+cu124
- Datasets: 3.0.1
- Tokenizers: 0.20.1
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
pictgensupport/infographicsv2 | pictgensupport | 2025-05-22T22:08:25Z | 0 | 0 | diffusers | [
"diffusers",
"flux",
"lora",
"replicate",
"text-to-image",
"en",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:other",
"region:us"
] | text-to-image | 2025-05-22T22:08:12Z | ---
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
language:
- en
tags:
- flux
- diffusers
- lora
- replicate
base_model: "black-forest-labs/FLUX.1-dev"
pipeline_tag: text-to-image
# widget:
# - text: >-
# prompt
# output:
# url: https://...
instance_prompt: ICON_BASIC
---
# Infographicsv2
<Gallery />
Trained on Replicate using:
https://replicate.com/ostris/flux-dev-lora-trainer/train
## Trigger words
You should use `ICON_BASIC` to trigger the image generation.
## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers)
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('pictgensupport/infographicsv2', weight_name='lora.safetensors')
image = pipeline('your prompt').images[0]
```
For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)
|
Dc-4nderson/results | Dc-4nderson | 2025-05-22T22:04:59Z | 7 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"bert",
"text-classification",
"generated_from_trainer",
"base_model:google-bert/bert-base-uncased",
"base_model:finetune:google-bert/bert-base-uncased",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | text-classification | 2025-04-17T22:47:17Z | ---
library_name: transformers
license: apache-2.0
base_model: bert-base-uncased
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: results
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# results
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0076
- Accuracy: 0.9984
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.1447 | 1.0 | 832 | 0.0211 | 0.9968 |
| 0.0152 | 2.0 | 1664 | 0.0230 | 0.9960 |
| 0.0184 | 3.0 | 2496 | 0.0118 | 0.9984 |
| 0.0074 | 4.0 | 3328 | 0.0089 | 0.9984 |
| 0.0098 | 5.0 | 4160 | 0.0076 | 0.9984 |
### Framework versions
- Transformers 4.52.3
- Pytorch 2.7.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1
|
CtrlAltArt/Flux_German_Film_Expressionism_Style | CtrlAltArt | 2025-05-22T22:02:51Z | 0 | 0 | diffusers | [
"diffusers",
"text-to-image",
"lora",
"template:diffusion-lora",
"base_model:black-forest-labs/FLUX.1-dev",
"base_model:adapter:black-forest-labs/FLUX.1-dev",
"license:mit",
"region:us"
] | text-to-image | 2025-05-22T22:02:07Z | ---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: >-
avt1vate! . Close up on a face with exaggerated, stark makeup, eyes wide
with terror, lit dramatically from below (chiaroscuro lighting). Style of
German Expressionism, intense psychological fear, monochrome
output:
url: images/ComfyUI_00033_.png
- text: >-
avt1vate! . A 'Femme Fatale' figure emerging from deep shadow, face
partially obscured, sharp lighting catching only an eye and starkly painted
lips. German Expressionist noir style, monochrome, mysterious and dangerous.
output:
url: images/ComfyUI_00050_.png
- text: >-
avt1vate! . Use of layered shadows: multiple figures or objects casting
overlapping, complex, and distorted shadows onto a stark wall or floor.
German Expressionism lighting technique, black and white, abstract
complexity.
output:
url: images/ComfyUI_00056_.png
- text: >-
A sleepwalker with stiff, unnatural movements navigates a crooked rooftop
landscape under a stark, artificial moon. German Expressionist aesthetic,
painted set look, black and white, dreamlike unease.
output:
url: images/ComfyUI_00034_.png
- text: >-
avt1vate! . Two figures in a tense, silent confrontation across a starkly
lit table in an angular room, their shadows elongated and distorted behind
them. German Expressionist drama, black and white, psychological intensity.
output:
url: images/ComfyUI_00041_.png
- text: >-
An empty, angular city street at night, buildings tilted precariously, sharp
geometric shadows, cobblestones lit by a single harsh light source. Style of
The Cabinet of Dr. Caligari, German Expressionism, black and white,
unsettling emptiness.
output:
url: images/ComfyUI_00036_.png
- text: >-
avt1vate! . A menacing shadow with elongated fingers creeping up a stark
white wall towards a terrified victim. Inspired by Nosferatu, German
Expressionism, high contrast lighting, intense suspense, monochrome.
output:
url: images/ComfyUI_00040_.png
- text: >-
avt1vate! . A crowd of figures with identical, mask-like faces moving
rigidly through a distorted, angular town square. German Expressionism
style, high contrast black and white, feeling of oppressive conformity.
output:
url: images/ComfyUI_00039_.png
- text: >-
avt1vate! . A character trapped behind sharply angled bars of shadow, face
pressed against them, eyes wide with desperation. German Expressionism, high
contrast monochrome, theme of imprisonment (psychological or physical).
output:
url: images/ComfyUI_00052_.png
- text: >-
avt1vate! . An angular, imposing courtroom scene: the judge's bench is a
towering, sharp-edged structure, figures cast long, distorted shadows, shot
from a low, unsettling angle. German Expressionism, monochrome, feeling of
judgment and doom.
output:
url: images/ComfyUI_00047_.png
- text: >-
A character hunched over, seemingly crushed by the weight of menacing,
leaning buildings on a narrow, dark street. Low angle shot, distorted
perspective, German Expressionism style, black and white.
output:
url: images/ComfyUI_00035_.png
- text: >-
avt1vate! . A distorted, nightmarish carnival scene: tents lean
precariously, carousel horses have grotesque faces, sharp shadows
everywhere. German Expressionist style, black and white, atmosphere of
sinister fun.
output:
url: images/ComfyUI_00044_.png
- text: >-
avt1vate! . A crowd surging through a warped street, faces blurred into
angular masks of collective emotion (fear, anger), lit by stark overhead
lamps. German Expressionism mob scene, black and white, loss of
individuality.
output:
url: images/ComfyUI_00053_.png
- text: >-
avt1vate! . A menacing, fog-shrouded harbor at night: docks twist at
impossible angles, ship masts form jagged lines against a pale sky, deep
shadows obscure the water. German Expressionist landscape, black and white,
eerie and isolating.
output:
url: images/ComfyUI_00046_.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: act1vate!
license: mit
---
# German Film Expressionism Style
<Gallery />
## Model description
Trigger words: does not need any but was trained with the word act1vate! at the start, to get as close to training data intended style as possible, include it.
recommended strength: 0.9-1 or higher, lower just makes most images look like a sketch but play with settings to get a feel for it.
Additional usage tips: After further testing I find using words like "a scene from a screen play", "This photograph shows" and other indications of real-life scenes, helps preventing the images looking like sketches.
Using terminology like "background made of cardboard", "warped and deformed stage set" helps recreate the stage backdrops in the correct style.
See my example images for full prompt examples and how to control the model.
## Trigger words
You should use `act1vate!` to trigger the image generation.
## Download model
Weights for this model are available in Safetensors format.
[Download](/CtrlAltArt/Flux_German_Film_Expressionism_Style/tree/main) them in the Files & versions tab.
|
ErasureResearch/esdx_tench | ErasureResearch | 2025-05-22T22:02:07Z | 0 | 0 | diffusers | [
"diffusers",
"safetensors",
"diffusion",
"concept-erasure",
"stable-diffusion",
"esdx",
"tench",
"text-to-image",
"en",
"dataset:imagenet",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | 2025-05-21T18:59:18Z | ---
license: mit
tags:
- diffusion
- concept-erasure
- stable-diffusion
- esdx
- tench
datasets:
- imagenet
language:
- en
pipeline_tag: text-to-image
---
# esdx_tench
This is a concept-erased Stable Diffusion model using the **Exact Source Distillation (ESD-X)** method to remove the concept **"Tench"**.
## Method
Exact Source Distillation (ESD-X) erases concepts by distilling knowledge while excluding specific concept representations.
## Usage
```python
from diffusers import StableDiffusionPipeline
import torch
pipe = StableDiffusionPipeline.from_pretrained("ErasureResearch/esdx_tench", torch_dtype=torch.float16).to("cuda")
prompt = "a photo of a tench"
image = pipe(prompt).images[0]
image.save("erased_tench.png")
```
## Citation
If you use this model in your research, please cite:
```bibtex
@article{concept_erasure_2024,
title={Concept Erasure in Diffusion Models},
author={ErasureResearch Team},
journal={Proceedings of...},
year={2024}
}
```
|
TheGardener/KD-Embedding-and-MLP-Llama-0.7B-epoch-3rd | TheGardener | 2025-05-22T21:57:06Z | 0 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | text-generation | 2025-05-22T21:56:28Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
Pentium95/h34v7_DXP-Zero-V1.0-24b-Small-iMatrix-GGUF | Pentium95 | 2025-05-22T21:56:22Z | 58 | 0 | null | [
"gguf",
"roleplay",
"storywriting",
"mistral",
"erp",
"imatrix",
"creative",
"creative writing",
"story",
"writing",
"roleplaying",
"role play",
"sillytavern",
"rp",
"text-generation",
"en",
"ru",
"base_model:h34v7/DXP-Zero-V1.0-24b-Small-Instruct",
"base_model:quantized:h34v7/DXP-Zero-V1.0-24b-Small-Instruct",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | text-generation | 2025-05-19T21:02:53Z | ---
license: apache-2.0
base_model:
- h34v7/DXP-Zero-V1.0-24b-Small-Instruct
base_model_relation: quantized
pipeline_tag: text-generation
tags:
- roleplay
- storywriting
- mistral
- erp
- gguf
- imatrix
- creative
- creative writing
- story
- writing
- roleplaying
- role play
- sillytavern
- rp
language:
- en
- ru
---
# Model Card for Model ID
Imatrix GGUF Quants for: [DXP-Zero-V1.0-24b-Small-Instruct](https://huggingface.co/h34v7/DXP-Zero-V1.0-24b-Small-Instruct#dxp-zero-v10-24b-small-instruct).
### Recommended Settings
```
"temperature": 0.8, (Mistral Small 3.1 is sensitive to higher temperatures)
"top_p": 0.95/1,
"min_p": 0.025/0.03,
"repeat_penalty": 1.05/1.1,
```
IQ2_M: Usable, good for 10-16 GB RAM/VRAM
IQ3_XXS: Very usable, good for 12-20 GB RAM/VRAM
IQ3_M: Solid, good for 14-18 GB RAM/VRAM
IQ4_XS: It's all you need, if you have 16+ GB RAM/VRAM
The model might lack the necessary evil for making story twisty or dark adventure but it make ammend on creating coherent story in long context form.
Perfect for romance, adventure, sci-fi, and even general purpose.
So i was browsing for Mistral finetune and found this base model by ZeroAgency, and oh boy... It was perfect!
So here are few notable improvements i observed. Pros:
Increased output for storytelling or roleplay.
Dynamic output (it can adjust how much output, i mean like when you go with shorter prompt it will do smaller outputs and so does with longer prompt more output too).
Less repetitive (though it depends on your own prompt and settings).
I have tested with 49444/65536 tokens no degradation although i notice it's actually learning the context better and it's impacting the output a lot. (what i don't like is, it's learning the previous context(of turns) too quickly and set it as new standards.).
This model was merged using the TIES merge method using ZeroAgency/Mistral-Small-3.1-24B-Instruct-2503-hf as a base. Models Merged:
PocketDoc/Dans-PersonalityEngine-V1.2.0-24b
Gryphe/Pantheon-RP-1.8-24b-Small-3.1 |
alexanderyj/gemma3_fine_tuning2025-05-22 | alexanderyj | 2025-05-22T21:54:29Z | 0 | 0 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:google/gemma-3-4b-it",
"base_model:finetune:google/gemma-3-4b-it",
"endpoints_compatible",
"region:us"
] | null | 2025-05-22T04:34:50Z | ---
base_model: google/gemma-3-4b-it
library_name: transformers
model_name: gemma3_fine_tuning2025-05-22
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for gemma3_fine_tuning2025-05-22
This model is a fine-tuned version of [google/gemma-3-4b-it](https://huggingface.co/google/gemma-3-4b-it).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="alexanderyj/gemma3_fine_tuning2025-05-22", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with SFT.
### Framework versions
- TRL: 0.17.0
- Transformers: 4.51.3
- Pytorch: 2.7.0
- Datasets: 3.6.0
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
``` |
cyberdelia/CyberRealisticPony | cyberdelia | 2025-05-22T21:54:20Z | 8,696 | 53 | diffusers | [
"diffusers",
"stable-diffusion",
"sdxl",
"text-to-image",
"photorealistic",
"cyberrealistic",
"pony",
"image-generation",
"license:creativeml-openrail-m",
"region:us"
] | text-to-image | 2024-05-09T10:12:22Z | ---
license: creativeml-openrail-m
tags:
- stable-diffusion
- sdxl
- text-to-image
- photorealistic
- cyberrealistic
- pony
- image-generation
- diffusers
model-index:
- name: CyberRealistic Pony
results: []
---
# CyberRealistic Pony
**CyberRealistic Pony** is the awesome Pony Diffusion with some CyberRealistic elements.
---
## ✨ Features
- **Photorealism**: Generates highly detailed and realistic pony images, capturing intricate textures and lighting.
- **Ease of Use**: Achieves impressive results with straightforward prompts.
- **Integrated VAE**: Comes with a baked-in Variational Autoencoder for enhanced image quality.
- **Versatility**: Suitable for various applications, including character design, illustrations, and concept art.
---
## 🛠️ Recommended Settings
| Parameter | Recommended Value |
|-----------------|------------------------------------------------|
| Sampling Steps | 30+ |
| Sampler | DPM++ SDE Karras / DPM++ 2M Karras / Euler a |
| Resolution | 896x1152 / 832x1216 |
| CFG Scale | 5 |
| VAE | Already baked-in |
---
## 🧾 Example Prompts
> score_9, score_8_up, score_7_up, (SUBJECT),
---
## 📸 Example Outputs
---
## 🔗 Links
- [Civitai Model Page](https://civitai.com/models/443821/cyberrealistic-pony)
---
## 🚫 Limitations
- May produce content that could be considered sensitive; use responsibly.
- Some prompts involving abstract or non-pony content may not perform as well due to the model's specialized training.
- Lighting and textures may occasionally be too clean or smooth depending on sampling choices.
---
## ✅ License
This model is distributed under the **CreativeML Open RAIL++-M License**, which allows commercial and non-commercial use, with proper credit and no malicious usage.
> [License details](https://huggingface.co/spaces/CompVis/stable-diffusion-license)
|
Subsets and Splits