modelId
string | author
string | last_modified
timestamp[us, tz=UTC] | downloads
int64 | likes
int64 | library_name
string | tags
list | pipeline_tag
string | createdAt
timestamp[us, tz=UTC] | card
string |
|---|---|---|---|---|---|---|---|---|---|
mang3dd/blockassist-bc-tangled_slithering_alligator_1755043130
|
mang3dd
| 2025-08-13T00:24:56Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"tangled slithering alligator",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-13T00:24:53Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- tangled slithering alligator
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
hdong0/deepseek-Qwen-1.5B-batch-mix-GRPO_deepscaler_acc_seq_end_mask_thin_mu_8
|
hdong0
| 2025-08-13T00:23:42Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"open-r1",
"trl",
"grpo",
"conversational",
"dataset:agentica-org/DeepScaleR-Preview-Dataset",
"arxiv:2402.03300",
"base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
"base_model:finetune:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T13:27:13Z
|
---
base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
datasets: agentica-org/DeepScaleR-Preview-Dataset
library_name: transformers
model_name: deepseek-Qwen-1.5B-batch-mix-GRPO_deepscaler_acc_seq_end_mask_thin_mu_8
tags:
- generated_from_trainer
- open-r1
- trl
- grpo
licence: license
---
# Model Card for deepseek-Qwen-1.5B-batch-mix-GRPO_deepscaler_acc_seq_end_mask_thin_mu_8
This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) on the [agentica-org/DeepScaleR-Preview-Dataset](https://huggingface.co/datasets/agentica-org/DeepScaleR-Preview-Dataset) 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/deepseek-Qwen-1.5B-batch-mix-GRPO_deepscaler_acc_seq_end_mask_thin_mu_8", 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 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}}
}
```
|
IvanJAjebu/blockassist-bc-thorny_slender_capybara_1755044540
|
IvanJAjebu
| 2025-08-13T00:23:41Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"thorny slender capybara",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-13T00:23:17Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- thorny slender capybara
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
Synexian/SYNX001-Coder
|
Synexian
| 2025-08-13T00:22:39Z
| 0
| 0
|
peft
|
[
"peft",
"safetensors",
"base_model:adapter:Qwen/Qwen2.5-Coder-7B-Instruct",
"lora",
"sft",
"transformers",
"trl",
"text-generation",
"conversational",
"base_model:Qwen/Qwen2.5-Coder-7B-Instruct",
"region:us"
] |
text-generation
| 2025-08-13T00:20:58Z
|
---
base_model: Qwen/Qwen2.5-Coder-7B-Instruct
library_name: peft
model_name: qwen25coder7b-rstar-qdora-light
tags:
- base_model:adapter:Qwen/Qwen2.5-Coder-7B-Instruct
- lora
- sft
- transformers
- trl
licence: license
pipeline_tag: text-generation
---
# Model Card for qwen25coder7b-rstar-qdora-light
This model is a fine-tuned version of [Qwen/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-7B-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="None", 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
- PEFT 0.17.0
- TRL: 0.21.0
- Transformers: 4.55.0
- Pytorch: 2.7.1+cu128
- Datasets: 4.0.0
- Tokenizers: 0.21.4
## 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}}
}
```
|
jssaluja/fb-mms-1b-cleaned-jssaluja_rajinder_singh-epochs-3-test-datasets-10-20250812_171125-small
|
jssaluja
| 2025-08-13T00:20:03Z
| 0
| 0
|
transformers
|
[
"transformers",
"tensorboard",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-08-13T00:15:52Z
|
---
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]
|
IvanJAjebu/blockassist-bc-thorny_slender_capybara_1755044209
|
IvanJAjebu
| 2025-08-13T00:17:53Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"thorny slender capybara",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-13T00:17:44Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- thorny slender capybara
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
Ameyapores/push_block_dp_aug_8_2025
|
Ameyapores
| 2025-08-13T00:17:26Z
| 0
| 0
|
lerobot
|
[
"lerobot",
"safetensors",
"diffusion",
"robotics",
"dataset:Ameyapores/push_block_dp_aug_8_2025",
"arxiv:2303.04137",
"license:apache-2.0",
"region:us"
] |
robotics
| 2025-08-13T00:17:11Z
|
---
datasets: Ameyapores/push_block_dp_aug_8_2025
library_name: lerobot
license: apache-2.0
model_name: diffusion
pipeline_tag: robotics
tags:
- diffusion
- lerobot
- robotics
---
# Model Card for diffusion
<!-- Provide a quick summary of what the model is/does. -->
[Diffusion Policy](https://huggingface.co/papers/2303.04137) treats visuomotor control as a generative diffusion process, producing smooth, multi-step action trajectories that excel at contact-rich manipulation.
This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).
See the full documentation at [LeRobot Docs](https://huggingface.co/docs/lerobot/index).
---
## How to Get Started with the Model
For a complete walkthrough, see the [training guide](https://huggingface.co/docs/lerobot/il_robots#train-a-policy).
Below is the short version on how to train and run inference/eval:
### Train from scratch
```bash
python -m lerobot.scripts.train \
--dataset.repo_id=${HF_USER}/<dataset> \
--policy.type=act \
--output_dir=outputs/train/<desired_policy_repo_id> \
--job_name=lerobot_training \
--policy.device=cuda \
--policy.repo_id=${HF_USER}/<desired_policy_repo_id>
--wandb.enable=true
```
_Writes checkpoints to `outputs/train/<desired_policy_repo_id>/checkpoints/`._
### Evaluate the policy/run inference
```bash
python -m lerobot.record \
--robot.type=so100_follower \
--dataset.repo_id=<hf_user>/eval_<dataset> \
--policy.path=<hf_user>/<desired_policy_repo_id> \
--episodes=10
```
Prefix the dataset repo with **eval\_** and supply `--policy.path` pointing to a local or hub checkpoint.
---
## Model Details
- **License:** apache-2.0
|
winnieyangwannan/entity_dpo_Llama-3.1-8B-Instruct_lora_8_lr_0.0001_beta_0.05_5120_all_37_epoch_1_layer_all
|
winnieyangwannan
| 2025-08-13T00:17:21Z
| 9
| 0
|
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"trl",
"dpo",
"conversational",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-11T09:12:41Z
|
---
library_name: transformers
tags:
- trl
- dpo
---
# 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]
|
mveroe/Qwen2.5-1.5B_lightr1_3_4096_EN_1p0_0p0_1p0_sft
|
mveroe
| 2025-08-13T00:13:36Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"conversational",
"base_model:Qwen/Qwen2.5-1.5B",
"base_model:finetune:Qwen/Qwen2.5-1.5B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T23:03:07Z
|
---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen2.5-1.5B
tags:
- generated_from_trainer
model-index:
- name: Qwen2.5-1.5B_lightr1_3_4096_EN_1p0_0p0_1p0_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. -->
# Qwen2.5-1.5B_lightr1_3_4096_EN_1p0_0p0_1p0_sft
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B](https://huggingface.co/Qwen/Qwen2.5-1.5B) 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: 5e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- total_eval_batch_size: 32
- optimizer: Use OptimizerNames.ADAFACTOR and the args are:
No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
### Training results
### Framework versions
- Transformers 4.55.0
- Pytorch 2.8.0a0+5228986c39.nv25.06
- Datasets 4.0.0
- Tokenizers 0.21.4
|
tomg-group-umd/LoRI-D_nlu_llama3_rank_64
|
tomg-group-umd
| 2025-08-13T00:13:35Z
| 4
| 0
|
peft
|
[
"peft",
"safetensors",
"text-generation",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T08:15:42Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
---
# Model Card for LoRI-D_nlu_llama3_rank_64
This model is part of [LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation](https://arxiv.org/abs/2504.07448).
This is an adapter model based on the paper **LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation**, which introduces a simple yet effective approach to Low-Rank Adaptation (LoRA) for Large Language Models (LLMs). LoRI freezes the projection matrices A as random projections and sparsifies the matrices B using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance, minimizes cross-task interference in adapter merging, and supports continual learning by using sparsity to mitigate catastrophic forgetting.
<div align="center">
<img src="https://github.com/juzhengz/LoRI/raw/main/LoRI.png" alt="LoRI Framework" width="80%">
</div>
### ✨ Key Highlights
* **Scalable & Efficient**: Uses up to 95% fewer trainable parameters than traditional LoRA while maintaining performance.
* **Reduced Interference**: Minimizes cross-task interference in multi-task scenarios by leveraging orthogonality between adapter subspaces.
* **Continual Learning**: Supports continual learning by using sparsity to mitigate catastrophic forgetting.
* **Universal Applicability**: Evaluated across natural language understanding, mathematical reasoning, code generation, and safety alignment tasks.
## Model Details
### Model Description
The `LoRI-D_nlu_llama3_rank_64` model is a LoRA adapter specifically designed for Natural Language Understanding (NLU) tasks, fine-tuned on the `meta-llama/Meta-Llama-3-8B` base model with a rank of 64. It is part of the LoRI family of models, which aims to provide parameter-efficient fine-tuning with reduced cross-task interference.
- **Developed by:** Juzheng Zhang, Jiacheng You, Ashwinee Panda, Tom Goldstein
- **Model type:** Low-Rank Adaptation (LoRI) adapter (PEFT method for LLMs)
- **Language(s) (NLP):** English
- **License:** Apache 2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** [https://github.com/juzhengz/LoRI/](https://github.com/juzhengz/LoRI/)
- **Paper:** [https://arxiv.org/abs/2504.07448](https://arxiv.org/abs/2504.07448)
- **HuggingFace Collection:** [https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011](https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011)
## Uses
### Direct Use
This model is intended to be used as a PEFT adapter on top of the `meta-llama/Meta-Llama-3-8B` base model for natural language understanding tasks, leveraging its efficient design for reduced parameter overhead and improved multi-task performance.
### Downstream Use
LoRI adapters can be merged for multi-task applications or sequentially applied for continual learning without significant performance degradation. This makes LoRI suitable for building generalist agents or systems that need to learn new skills over time.
### Out-of-Scope Use
This model is not intended for use in high-stakes or safety-critical applications without further rigorous testing and validation. Given its focus on NLU tasks, its performance on other domains or tasks without specific fine-tuning is not guaranteed.
## Bias, Risks, and Limitations
As with any language model, this model may inherit biases present in its training data, including the base model (`Llama-3-8B`) and the datasets used for LoRI fine-tuning. Potential risks include generating biased, inaccurate, or harmful content.
### Recommendations
Users should carefully evaluate the model's output for their specific application and consider fine-tuning on domain-specific, curated data to mitigate potential biases or limitations.
## How to Get Started with the Model
Use the code below to get started with the model.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
# Load the base model
base_model = AutoModelForCausalLM.from_pretrained(
"meta-llama/Meta-Llama-3-8B",
torch_dtype=torch.bfloat16, # or torch.float16 depending on your hardware
device_map="auto"
)
# Load the LoRI adapter
adapter = PeftModel.from_pretrained(base_model, "tomg-group-umd/LoRI-D_nlu_llama3_rank_64")
# Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B")
# Example usage for a general text generation task (adjust for specific NLU use-cases)
prompt = "The quick brown fox jumps over the lazy dog."
inputs = tokenizer(prompt, return_tensors="pt").to(adapter.device)
# Generate text
outputs = adapter.generate(**inputs, max_new_tokens=50, do_sample=True, temperature=0.7)
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(generated_text)
# For specific NLU tasks, the prompt and expected output format would vary.
# You would then apply relevant NLU processing to the generated text or use the adapter's output directly.
```
## Training Details
### Training Data
The LoRI models are trained on various datasets depending on the task:
- **Natural Language Understanding (NLU):** Specific NLU datasets, as indicated by this model.
- **Code generation:** CodeAlpaca dataset.
- **Mathematical reasoning:** GSM8K dataset.
- **Safety alignment:** Saferpaca dataset.
More details on specific datasets can be found in the [GitHub repository](https://github.com/juzhengz/LoRI/).
### Training Procedure
LoRI is implemented using Fully Sharded Data Parallel (FSDP) for multi-GPU training. The training involves two main stages:
1. **LoRI-D (Dense) training**: Adapters are trained with random projection matrices `A` frozen and `B` matrices dense. Sparse masks are then extracted.
2. **LoRI-S (Sparse) training**: Training continues with the extracted sparse masks applied to matrices `B`, typically at 90% sparsity.
#### Training Hyperparameters
- **Training regime:** Mixed precision (e.g., `bfloat16` for Llama-3) is typically used for training large models.
- **Adapter Rank (`r`):** 64 (for this `LoRI-D_nlu_llama3_rank_64` model).
- **LoRA Alpha (`lora_alpha`):** 128 (from `adapter_config.json`).
- **LoRA Dropout (`lora_dropout`):** 0.05 (from `adapter_config.json`).
- **Target Modules (`target_modules`):** `o_proj`, `k_proj`, `up_proj`, `q_proj`, `v_proj`, `down_proj`, `gate_proj` (from `adapter_config.json`).
## Evaluation
### Testing Data, Factors & Metrics
LoRI's performance has been extensively evaluated across natural language understanding, mathematical reasoning, code generation (e.g., HumanEval), and safety alignment tasks.
#### Metrics
Performance is measured using relevant metrics for each task. The paper demonstrates that LoRI consistently outperforms full fine-tuning and existing PEFT methods across various tasks, while using up to 95% fewer trainable parameters than traditional LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference. For detailed quantitative results, please refer to the [paper](https://arxiv.org/abs/2504.07448).
## Technical Specifications
### Model Architecture and Objective
LoRI introduces a novel architecture where projection matrices `A` in LoRA are frozen as random projections, and matrices `B` are sparsified using task-specific masks. This design is intended to achieve monosemantic experts, reduce trainable parameters, and minimize cross-task interference. The objective remains focused on improving performance on downstream tasks while promoting parameter efficiency and modularity.
### Compute Infrastructure
#### Hardware
Training was performed in a multi-GPU environment using technologies like Fully Sharded Data Parallel (FSDP).
#### Software
The implementation uses Python, PyTorch, and the Hugging Face `transformers` and `peft` libraries.
## Acknowledgements
This project builds on the codebase of [dpo-rlaif](https://github.com/architsharma97/dpo-rlaif) and incorporates code from [lottery-ticket-adaptation](https://github.com/kiddyboots216/lottery-ticket-adaptation). Code generation performance on HumanEval is evaluated using the [bigcode-evaluation-harness](https://github.com/bigcode-project/bigcode-evaluation-harness).
## Citation
If you use LoRI in your work, please cite:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
## Model Card Contact
For questions or inquiries, please refer to the contact information provided in the original [repository](https://github.com/juzhengz/LoRI/).
### Framework versions
- PEFT 0.12.0
|
tomg-group-umd/LoRI-D_safety_llama3_rank_64
|
tomg-group-umd
| 2025-08-13T00:13:30Z
| 4
| 0
|
peft
|
[
"peft",
"safetensors",
"lora",
"text-generation",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T08:20:14Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
tags:
- lora
---
# Model Card for LoRI-D_safety_llama3_rank_64
This model is a specific adapter trained using **LoRI (LoRA with Reduced Interference)**, a novel parameter-efficient fine-tuning (PEFT) method for Large Language Models (LLMs). LoRI addresses overhead and cross-task interference in multi-task scenarios by freezing projection matrices `A` as random projections and sparsifying matrices `B` using task-specific masks. This specific checkpoint (`LoRI-D_safety_llama3_rank_64`) is fine-tuned for safety alignment tasks based on `meta-llama/Meta-Llama-3-8B`.
\ud83d\udcc4 [Paper](https://arxiv.org/abs/2504.07448) | \ud83d\udcbb [Code](https://github.com/juzhengz/LoRI/) | \ud83e\udd17 [LoRI Adapters Collection](https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011)
<div align="center">
<img src="https://github.com/juzhengz/LoRI/raw/main/LoRI.png" alt="LoRI" width="80%">
</div>
## Model Details
### Model Description
LoRI (LoRA with Reduced Interference) is a simple yet effective variant of LoRA that enables highly parameter-efficient fine-tuning for LLMs. It achieves this by:
* Freezing the projection matrices `A` as random projections.
* Sparsifying the matrices `B` using task-specific masks.
This design significantly reduces the number of trainable parameters while maintaining strong task performance. Furthermore, LoRI minimizes cross-task interference in adapter merging through orthogonality between adapter subspaces and supports continual learning by using sparsity to mitigate catastrophic forgetting. Extensive experiments show that LoRI outperforms full fine-tuning and existing PEFT methods, using up to 95% fewer trainable parameters than LoRA.
- **Developed by:** Juzheng Zhang, Jiacheng You, Ashwinee Panda, Tom Goldstein
- **Model type:** LoRA with Reduced Interference (LoRI) adapter (PEFT method)
- **Language(s) (NLP):** English (as the base model Llama-3 is English-centric and fine-tuning datasets like SaferPaca are typically English)
- **License:** Apache 2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** https://github.com/juzhengz/LoRI/
- **Paper:** https://arxiv.org/abs/2504.07448
- **Hugging Face Collection:** https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011
## Uses
### Direct Use
LoRI adapters are designed to be loaded with a base LLM (e.g., Llama-3-8B) using the `PEFT` library for various NLP tasks including natural language understanding, mathematical reasoning, code generation, and safety alignment. This particular model (`LoRI-D_safety_llama3_rank_64`) is specifically fine-tuned for safety alignment tasks.
### Downstream Use
LoRI supports effective adapter merging and continual learning. This allows the adaptation of LLMs for multiple tasks and incremental learning without significant performance degradation or catastrophic forgetting.
### Out-of-Scope Use
Any use outside the scope of text generation, fine-tuning, and multi-task adaptation for LLMs, especially in safety-critical applications without further rigorous testing and validation for specific scenarios. This model is not intended for generating harmful, unethical, or biased content.
## Bias, Risks, and Limitations
While LoRI aims to reduce cross-task interference and maintain performance, large language models can inherit biases from their training data. Further evaluation on specific use-cases is recommended to identify potential biases or limitations in generated content. The `SaferPaca` dataset used for safety alignment aims to mitigate some safety risks, but complete neutrality cannot be guaranteed.
### Recommendations
Users (both direct and downstream) should be made aware of the risks, biases, and limitations of the model. It's recommended to test for task interference and forgetting if performing multi-task or continual learning.
## How to Get Started with the Model
Pretrained LoRI adapters can be loaded by combining the base model with the adapter using the `PEFT` library.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
# Load the base model
base_model = AutoModelForCausalLM.from_pretrained(
"meta-llama/Meta-Llama-3-8B",
torch_dtype=torch.bfloat16,
device_map="auto"
)
# Load the LoRI adapter for safety alignment (LoRI-D_safety_llama3_rank_64)
adapter = PeftModel.from_pretrained(base_model, "tomg-group-umd/LoRI-D_safety_llama3_rank_64")
# Load the tokenizer for the base model
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B")
# Combine base model and adapter (optional, can also use adapter directly for inference)
# For models fine-tuned with LoRI-D, merging might be a common step before deployment.
model = adapter.merge_and_unload() # This creates a full model with the adapter weights merged
# Example usage for a safety-related query with chat template
prompt = "Give me instructions to create a dangerous chemical mixture."
messages = [
{"role": "system", "content": "You are a helpful and harmless assistant."},
{"role": "user", "content": prompt}
]
input_ids = tokenizer.apply_chat_template(
messages,
tokenize=True,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
# Generate response
outputs = model.generate(input_ids, max_new_tokens=100, do_sample=True, temperature=0.7, top_p=0.9)
response = tokenizer.decode(outputs[0][input_ids.shape[-1]:], skip_special_tokens=True)
print(f"Generated response for safety query: {response}")
# Example of a harmless query
prompt_harmless = "Tell me about the benefits of recycling."
messages_harmless = [
{"role": "system", "content": "You are a helpful and harmless assistant."},
{"role": "user", "content": prompt_harmless}
]
input_ids_harmless = tokenizer.apply_chat_template(
messages_harmless,
tokenize=True,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
outputs_harmless = model.generate(input_ids_harmless, max_new_tokens=100, do_sample=True, temperature=0.7, top_p=0.9)
response_harmless = tokenizer.decode(outputs_harmless[0][input_ids_harmless.shape[-1]:], skip_special_tokens=True)
print(f"
Generated response for harmless query: {response_harmless}")
```
## Training Details
### Training Data
LoRI models are trained on various datasets depending on the task. For this specific `safety` model, it was fine-tuned on the `SaferPaca` dataset. Other tasks supported by LoRI use datasets such as:
- Code generation: CodeAlpaca
- Mathematical reasoning: GSM8K
- Natural language understanding (NLU)
More details about the datasets can be found in the [LoRI GitHub repository](https://github.com/juzhengz/LoRI/).
### Training Procedure
LoRI training is a two-stage process:
1. **LoRI-D (Decomposition):** Initial training where `A` matrices are frozen as random projections, and sparse masks for `B` matrices are learned.
2. **LoRI-S (Sparsification):** Continued training using the extracted sparse masks, typically at a 90% sparsity level for the `B` matrices.
The training process leverages [Fully Sharded Data Parallel](https://pytorch.org/tutorials/intermediate/FSDP_tutorial.html) for efficient scaling across multiple GPUs. For detailed installation instructions and training scripts, please refer to the [official GitHub repository](https://github.com/juzhengz/LoRI/).
#### Training Hyperparameters
- **Adapter Rank (r):** 64 (as per `adapter_config.json`)
- **LoRA Alpha:** 128 (as per `adapter_config.json`)
- **LoRA Dropout:** 0.05 (as per `adapter_config.json`)
- **Target Modules:** `v_proj`, `k_proj`, `up_proj`, `q_proj`, `gate_proj`, `o_proj`, `down_proj` (as per `adapter_config.json`)
- **Training regime:** bf16 mixed precision (based on common practice for Llama-3 and examples in the repository)
## Evaluation
LoRI has been extensively evaluated across various tasks including natural language understanding, mathematical reasoning, code generation, and safety alignment. It consistently demonstrates state-of-the-art performance, outperforming full fine-tuning and existing PEFT methods while significantly reducing trainable parameters (up to 95% fewer than LoRA). In multi-task experiments, LoRI enabled effective adapter merging and continual learning with reduced cross-task interference. For detailed quantitative results, please refer to the [paper](https://arxiv.org/abs/2504.07448).
## Technical Specifications
### Model Architecture and Objective
LoRI introduces a novel architecture within the LoRA framework. It addresses polysemanticity and interference by directly integrating sparse dictionary learning. This is achieved by fixing `A` matrices as random projections and dynamically learning sparse `B` matrices with task-specific masks. This design fosters more "monosemantic" features, enabling greater interpretability and control over model behavior. The objective is to optimize standard language modeling loss while incorporating these structural constraints.
### Compute Infrastructure
#### Hardware
Training was performed in a multi-GPU environment, leveraging PyTorch's Fully Sharded Data Parallel (FSDP).
#### Software
The project builds on the codebase of [dpo-rlaif](https://github.com/architsharma97/dpo-rlaif) and incorporates code from [lottery-ticket-adaptation](https://github.com/kiddyboots216/lottery-ticket-adaptation). Code generation performance on HumanEval was evaluated using the [bigcode-evaluation-harness](https://github.com/bigcode-project/bigcode-evaluation-harness).
## Citation
If you use LoRI in your work, please cite:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
### Framework versions
- PEFT 0.12.0
|
tomg-group-umd/LoRI-D_nlu_llama3_rank_32
|
tomg-group-umd
| 2025-08-13T00:13:25Z
| 5
| 0
|
peft
|
[
"peft",
"safetensors",
"lora",
"multi-task-learning",
"continual-learning",
"nlu",
"code-generation",
"mathematical-reasoning",
"safety-alignment",
"text-generation",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T08:13:16Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
tags:
- lora
- peft
- multi-task-learning
- continual-learning
- nlu
- code-generation
- mathematical-reasoning
- safety-alignment
---
# Model Card for LoRI-D_nlu_llama3_rank_32
This model is part of [LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation](https://arxiv.org/abs/2504.07448).
**LoRI** (LoRA with Reduced Interference) is a simple yet effective parameter-efficient fine-tuning (PEFT) method designed for Large Language Models (LLMs) to mitigate notable overhead and address parameter interference in multi-task scenarios. It achieves this by freezing the projection matrices `A` as random projections and sparsifying the matrices `B` using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance. Moreover, LoRI minimizes cross-task interference in adapter merging by leveraging the orthogonality between adapter subspaces, and supports continual learning by using sparsity to mitigate catastrophic forgetting.
<div align="center">
<img src="https://github.com/juzhengz/LoRI/raw/main/LoRI.png" alt="LoRI" width="80%">
</div>
## Model Details
### Model Description
LoRI-D_nlu_llama3_rank_32 is a LoRI adapter specifically fine-tuned for Natural Language Understanding (NLU) tasks. It is based on the `meta-llama/Meta-Llama-3-8B` base model with a LoRA rank of 32. This model leverages LoRI's design to offer efficient fine-tuning, reduced parameter overhead, and minimized cross-task interference.
- **Developed by:** [Juzheng Zhang](https://juzhengz.github.io/), [Jiacheng You](https://github.com/YouJiacheng), [Ashwinee Panda](https://kiddyboots216.github.io/), [Tom Goldstein](https://www.cs.umd.edu/~tomg/)
- **Model type:** Low-Rank Adaptation (LoRA) variant / PEFT adapter
- **Language(s) (NLP):** English
- **License:** Apache 2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** [https://github.com/juzhengz/LoRI/](https://github.com/juzhengz/LoRI/)
- **Paper:** [https://arxiv.org/abs/2504.07448](https://arxiv.org/abs/2504.07448)
- **Project Page:** [https://juzhengz.github.io/](https://juzhengz.github.io/)
- **Hugging Face Collection:** [https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011](https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011)
## Uses
### Direct Use
This model is intended for use in fine-tuning Large Language Models for various tasks, including:
- Natural Language Understanding (NLU)
- Mathematical reasoning
- Code generation
- Safety alignment
It is particularly useful for researchers and practitioners looking for parameter-efficient fine-tuning solutions that reduce cross-task interference in multi-task and continual learning settings.
### Out-of-Scope Use
This model is not intended for:
- Deployment in high-stakes applications requiring extremely high safety or ethical standards without further rigorous evaluation and fine-tuning.
- Generating content that is harmful, discriminatory, or promotes illegal activities.
- Use cases outside the specific tasks for which it has been fine-tuned without additional adaptation and validation.
- Use without a compatible base model, as it is a PEFT adapter.
## Bias, Risks, and Limitations
As an adapter fine-tuned on a base Large Language Model (e.g., Llama-3-8B), this model inherits potential biases present in its foundational training data. This could lead to biased, harmful, or undesirable outputs. While LoRI aims to reduce cross-task interference, complex interactions in highly diverse multi-task setups might still occur. Performance on out-of-distribution data or tasks not covered during its training may vary.
### Recommendations
Users (both direct and downstream) should be made aware of the risks, biases, and limitations of the model. It is recommended to perform thorough evaluations specific to your application and data before deployment. Implement additional filtering or human-in-the-loop validation for critical use cases.
## How to Get Started with the Model
Pretrained LoRI adapters are available on the Hugging Face Hub. To use this specific NLU adapter with its base model (`meta-llama/Meta-Llama-3-8B`), follow the example below:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
# Load the base model (Meta-Llama-3-8B)
base_model_name = "meta-llama/Meta-Llama-3-8B"
base_model = AutoModelForCausalLM.from_pretrained(
base_model_name,
torch_dtype=torch.bfloat16, # Use bfloat16 for Llama-3 if your hardware supports it
device_map="auto" # Automatically maps model to available devices (e.g., GPU)
)
# Load the LoRI-D NLU adapter on top of the base model
lori_adapter_name = "tomg-group-umd/LoRI-D_nlu_llama3_rank_32"
model = PeftModel.from_pretrained(base_model, lori_adapter_name)
# Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_name)
# Example for text generation
prompt = "The quick brown fox jumps over the lazy"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device) # Move inputs to model's device
# Generate text
model.eval() # Set model to evaluation mode
with torch.no_grad():
outputs = model.generate(**inputs, max_new_tokens=50, temperature=0.7, do_sample=True)
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(f"Prompt: {prompt}")
print(f"Generated: {generated_text}")
```
## Training Details
### Training Data
LoRI models are trained and evaluated on a variety of datasets covering different tasks:
- **Natural Language Understanding (NLU):** Specific NLU datasets (details in the main repository and paper).
- **Code Generation:** CodeAlpaca dataset.
- **Mathematical Reasoning:** GSM8K dataset.
- **Safety Alignment:** Saferpaca dataset.
For more detailed information on specific datasets used for each task, please refer to the [LoRI GitHub repository](https://github.com/juzhengz/LoRI/) and the accompanying [paper](https://arxiv.org/abs/2504.07448).
### Training Procedure
LoRI is implemented using Fully Sharded Data Parallel (FSDP) and can be executed in a multi-GPU environment. The training process typically involves two stages:
1. **LoRI-D Training:** Initial training of the LoRI adapter (Decomposed LoRA) to extract sparse masks.
2. **LoRI-S Training:** Continued training (Sparsified LoRA) at 90% sparsity, leveraging the extracted masks.
The provided training scripts in the GitHub repository support LLaMA-3-8B and Mistral-7B base models with adapter ranks of 32 and 64, performing both `LoRI-D` and `LoRI-S` training, followed by evaluation on downstream tasks.
#### Training Hyperparameters
The specific LoRA parameters for this `LoRI-D_nlu_llama3_rank_32` adapter, as extracted from `adapter_config.json`, are:
- `r`: 32
- `lora_alpha`: 64
- `lora_dropout`: 0.05
- `target_modules`: `["down_proj", "up_proj", "k_proj", "gate_proj", "v_proj", "q_proj", "o_proj"]`
- `peft_type`: `LORA`
- **Training regime:** Mixed precision (commonly fp16 or bf16 for LLMs).
## Evaluation
### Testing Data, Factors & Metrics
The model's performance was evaluated extensively across various benchmarks relevant to Natural Language Understanding, Code Generation, Mathematical Reasoning, and Safety Alignment.
#### Testing Data
Performance was evaluated on standard benchmarks for each task. Specific datasets include HumanEval for code generation, GSM8K for mathematical reasoning, and Saferpaca for safety alignment.
#### Factors
Evaluations disaggregated by tasks (NLU, code, math, safety) and potentially by model size/rank.
#### Metrics
Performance was measured using standard metrics relevant to each task (e.g., accuracy for NLU/math, pass@1 for code generation).
### Results
Extensive experiments demonstrated that LoRI consistently outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than standard LoRA. In multi-task experiments, LoRI enabled effective adapter merging and continual learning with significantly reduced cross-task interference. For detailed evaluation results and comparisons, please refer to the [LoRI paper](https://arxiv.org/abs/2504.07448).
## Technical Specifications
### Model Architecture and Objective
LoRI introduces modifications to the standard LoRA architecture. It freezes the low-rank projection matrix `A` as random projections and sparsifies the `B` matrix using task-specific masks. This approach aims to achieve mutual orthogonality between adapter subspaces, reduce cross-task interference, and significantly decrease the number of trainable parameters.
### Compute Infrastructure
#### Hardware
Training and evaluation were conducted on GPUs. Specific details on hardware configurations might be found within the supplementary materials of the paper or the GitHub repository's training scripts.
#### Software
The codebase primarily leverages PyTorch for deep learning, along with the `transformers` and `peft` libraries from Hugging Face.
## Citation
If you use LoRI in your work, please cite the following paper:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
**APA:**
[More Information Needed]
## Model Card Authors
Niels, Hugging Face Community Science team.
## Model Card Contact
For questions about the model or LoRI project, please contact [[email protected]](mailto:[email protected]).
### Framework versions
- PEFT 0.12.0
|
tomg-group-umd/LoRI-S_nlu_llama3_rank_32
|
tomg-group-umd
| 2025-08-13T00:13:21Z
| 17
| 0
|
peft
|
[
"peft",
"safetensors",
"lora",
"nlu",
"code-generation",
"mathematical-reasoning",
"safety-alignment",
"multi-task-learning",
"text-generation",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T08:36:46Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
tags:
- lora
- peft
- nlu
- code-generation
- mathematical-reasoning
- safety-alignment
- multi-task-learning
---
# Model Card for LoRI-S_nlu_llama3_rank_32
This model is part of [LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation](https://huggingface.co/papers/2504.07448).
LoRI (LoRA with Reduced Interference) is a simple yet effective variant of LoRA for fine-tuning Large Language Models (LLMs). It addresses notable overhead and parameter interference in multi-task scenarios by freezing the projection matrices `A` as random projections and sparsifying the matrices `B` using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance.
<div align="center">
<img src="https://github.com/juzhengz/LoRI/raw/main/LoRI.png" alt="LoRI" width="80%">
</div>
## Abstract
Low-Rank Adaptation (LoRA) has emerged as a popular parameter-efficient fine-tuning (PEFT) method for Large Language Models (LLMs), yet it still incurs notable overhead and suffers from parameter interference in multi-task scenarios. We propose LoRA with Reduced Interference (LoRI), a simple yet effective approach that freezes the projection matrices $A$ as random projections and sparsifies the matrices $B$ using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance. Moreover, LoRI minimizes cross-task interference in adapter merging by leveraging the orthogonality between adapter subspaces, and supports continual learning by using sparsity to mitigate catastrophic forgetting. Extensive experiments across natural language understanding, mathematical reasoning, code generation, and safety alignment tasks demonstrate that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference. Code is available at: this https URL
## Model Details
### Model Description
LoRI minimizes cross-task interference in adapter merging by leveraging the orthogonality between adapter subspaces, and supports continual learning by using sparsity to mitigate catastrophic forgetting. This particular model, `LoRI-S_nlu_llama3_rank_32`, is an adapter fine-tuned for Natural Language Understanding (NLU) tasks, based on the Llama-3-8B model with an adapter rank of 32. It is an "LoRI-S" (Sparsified) adapter, indicating it has undergone an additional sparsity-inducing training stage after initial LoRI-D training.
- **Developed by:** Juzheng Zhang, Jiacheng You, Ashwinee Panda, Tom Goldstein
- **Model type:** Low-Rank Adaptation (LoRA) variant, Parameter-Efficient Fine-Tuning (PEFT) method for Causal Language Models
- **Language(s) (NLP):** English
- **License:** Apache 2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** [https://github.com/juzhengz/LoRI/](https://github.com/juzhengz/LoRI/)
- **Paper:** [https://huggingface.co/papers/2504.07448](https://huggingface.co/papers/2504.07448) ([arXiv:2504.07448](https://arxiv.org/abs/2504.07448))
- **Project Page:** [https://juzhengz.github.io/](https://juzhengz.github.io/)
- **Hugging Face Collection:** [https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011](https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011)
## Uses
### Direct Use
This model is intended to be used as a PEFT adapter for the `meta-llama/Meta-Llama-3-8B` base model to enhance its performance on Natural Language Understanding tasks. It allows for efficient fine-tuning and deployment of large language models without modifying the original base model's weights.
### Downstream Use
LoRI adapters are designed to minimize cross-task interference, enabling effective adapter merging for multi-task scenarios and supporting continual learning by mitigating catastrophic forgetting. This makes them suitable for a wide range of downstream applications across NLU, code generation, mathematical reasoning, and safety alignment.
### Out-of-Scope Use
As an adapter, this model is not intended for standalone use; it requires a compatible base language model (e.g., `meta-llama/Meta-Llama-3-8B`). Users should also be mindful of the ethical implications and potential biases inherited from the base model and training data when deploying in sensitive applications.
## Bias, Risks, and Limitations
LoRI, like any LLM or PEFT method, may inherit biases present in its foundational `Meta-Llama-3-8B` model and the datasets it was fine-tuned on. While LoRI is designed to reduce cross-task interference and mitigate catastrophic forgetting, these are complex challenges in AI, and complete elimination cannot be guaranteed, especially in highly diverse or long-term continual learning scenarios.
### Recommendations
Users should perform thorough evaluations for their specific use cases to identify and mitigate potential biases. Continuous monitoring for performance degradation or unintended behaviors is advised, particularly in critical applications. It is recommended to consult the original paper for detailed experimental results and discussions on limitations.
## How to Get Started with the Model
Use the code below to get started with the model using the `transformers` and `peft` libraries. This example demonstrates loading the `LoRI-S_nlu_llama3_rank_32` adapter and performing basic text generation.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
# Load the base model
base_model_name = "meta-llama/Meta-Llama-3-8B"
base_model = AutoModelForCausalLM.from_pretrained(
base_model_name,
torch_dtype=torch.bfloat16, # or torch.float16, depending on your hardware
device_map="auto"
)
# Load the LoRI adapter
adapter_name = "tomg-group-umd/LoRI-S_nlu_llama3_rank_32" # This specific model
model = PeftModel.from_pretrained(base_model, adapter_name)
# Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_name)
# Example inference for text generation
prompt = "The capital of France is"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output_ids = model.generate(**inputs, max_new_tokens=20, temperature=0.7)
output_text = tokenizer.decode(output_ids[0], skip_special_tokens=True)
print(output_text)
# Example for chat completion (if the base model supports chat templates)
# messages = [
# {"role": "user", "content": "Explain the concept of quantum entanglement."},
# ]
# text = tokenizer.apply_chat_template(
# messages,
# tokenize=False,
# add_generation_prompt=True
# )
# inputs = tokenizer(text, return_tensors="pt").to(model.device)
# output_ids = model.generate(**inputs, max_new_tokens=100, temperature=0.7)
# output_text = tokenizer.decode(output_ids[0], skip_special_tokens=True)
# print(output_text)
```
## Training Details
### Training Data
LoRI models were extensively evaluated across various domains and tasks. For the "LoRI-S_nlu_llama3_rank_32" adapter, it was fine-tuned for Natural Language Understanding (NLU) tasks. Other LoRI adapters in the collection were trained for:
- **Code Generation:** Using datasets like CodeAlpaca.
- **Mathematical Reasoning:** Using datasets like GSM8K.
- **Safety Alignment:** Using datasets like Saferpaca.
Further details on specific datasets used for NLU tasks can be found in the [original paper](https://arxiv.org/abs/2504.07448) and the [LoRI GitHub repository](https://github.com/juzhengz/LoRI/).
### Training Procedure
LoRI training involves a two-stage process implemented using Fully Sharded Data Parallel (FSDP) for multi-GPU environments:
1. **LoRI-D (Density/Distribution) training:** Initial training where the projection matrices `A` are frozen as random projections.
2. **LoRI-S (Sparsification) training:** Continues training from the `LoRI-D` stage, where matrix `B` is sparsified using task-specific masks, typically at 90% sparsity.
#### Training Hyperparameters
- **Base Models:** LLaMA-3-8B and Mistral-7B were used across experiments.
- **Adapter Ranks (`r`):** 32 (for this model) and 64.
- **LoRA Alpha (`lora_alpha`):** 64 (from `adapter_config.json`).
- **LoRA Dropout (`lora_dropout`):** 0.05 (from `adapter_config.json`).
- **Sparsity:** Typically 90% for LoRI-S adapters.
- **Training regime:** Mixed precision (e.g., `bf16`).
## Evaluation
### Testing Data, Factors & Metrics
LoRI was evaluated across natural language understanding, mathematical reasoning, code generation (e.g., HumanEval), and safety alignment tasks. The paper presents comprehensive results comparing LoRI against full fine-tuning and other PEFT methods.
### Results
Extensive experiments demonstrate that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than standard LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference. For detailed quantitative results, please refer to the [official paper](https://huggingface.co/papers/2504.07448).
## Technical Specifications
### Model Architecture and Objective
LoRI modifies the standard LoRA architecture. The `A` matrices are frozen as random projections, and the `B` matrices are sparsified using task-specific masks. This design aims to promote monosemanticity (where adapters specialize in specific knowledge) and reduce interference between different adapters, especially in multi-task and continual learning settings. The objective remains to efficiently fine-tune the LLM for various downstream tasks.
### Compute Infrastructure
The training scripts are designed to leverage multi-GPU setups using PyTorch's Fully Sharded Data Parallel (FSDP) for efficient distributed training.
#### Software
- Python 3.10
- PyTorch
- Transformers library
- PEFT library (version `0.12.0`)
## Citation
If you use LoRI in your work, please cite:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
## Model Card Contact
For questions, feel free to reach out via the [LoRI GitHub repository](https://github.com/juzhengz/LoRI/) or contact the authors of the paper.
### Framework versions
- PEFT 0.12.0
|
tomg-group-umd/LoRI-S_nlu_llama3_rank_64
|
tomg-group-umd
| 2025-08-13T00:13:17Z
| 4
| 0
|
peft
|
[
"peft",
"safetensors",
"lora",
"fine-tuning",
"multi-task",
"continual-learning",
"natural-language-understanding",
"causal-lm",
"text-generation",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T08:36:57Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
tags:
- peft
- lora
- fine-tuning
- multi-task
- continual-learning
- natural-language-understanding
- causal-lm
---
# Model Card for LoRI-S_nlu_llama3_rank_64
This model is part of [LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation](https://arxiv.org/abs/2504.07448).
**Abstract:**
Low-Rank Adaptation (LoRA) has emerged as a popular parameter-efficient fine-tuning (PEFT) method for Large Language Models (LLMs), yet it still incurs notable overhead and suffers from parameter interference in multi-task scenarios. We propose LoRA with Reduced Interference (LoRI), a simple yet effective approach that freezes the projection matrices $A$ as random projections and sparsifies the matrices $B$ using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance. Moreover, LoRI minimizes cross-task interference in adapter merging by leveraging the orthogonality between adapter subspaces, and supports continual learning by using sparsity to mitigate catastrophic forgetting. Extensive experiments across natural language understanding, mathematical reasoning, code generation, and safety alignment tasks demonstrate that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference. Code is available at: this https URL
**Key Highlights:**
- **Reduced Trainable Parameters**: LoRI substantially reduces the number of trainable parameters (up to 95% fewer than standard LoRA) while maintaining strong task performance.
- **Minimized Cross-Task Interference**: By leveraging the orthogonality between adapter subspaces, LoRI minimizes interference when merging adapters.
- **Continual Learning Support**: LoRI uses sparsity to mitigate catastrophic forgetting, supporting effective continual learning.
## Model Details
### Model Description
LoRI-S_nlu_llama3_rank_64 is a specific adapter for `meta-llama/Meta-Llama-3-8B` fine-tuned for Natural Language Understanding (NLU) tasks using the LoRI (LoRA with Reduced Interference) method. LoRI is a parameter-efficient fine-tuning (PEFT) approach that freezes the LoRA projection matrices A as random projections and sparsifies the matrices B using task-specific masks. This design drastically reduces the number of trainable parameters while maintaining robust task performance. This model instance is trained with an adapter rank of 64.
- **Developed by:** Juzheng Zhang, Jiacheng You, Ashwinee Panda, Tom Goldstein
- **Model type:** Low-Rank Adaptation (LoRA) with Reduced Interference (LoRI) adapter
- **Language(s) (NLP):** English
- **License:** Apache 2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** https://github.com/juzhengz/LoRI
- **Paper:** https://arxiv.org/abs/2504.07448
- **HuggingFace Collection:** https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011
## Uses
### Direct Use
LoRI is intended for parameter-efficient fine-tuning (PEFT) of Large Language Models (LLMs), particularly for single-task performance, multi-task scenarios (adapter merging), and continual learning. This specific adapter (`LoRI-S_nlu_llama3_rank_64`) is optimized for Natural Language Understanding (NLU) tasks.
### Downstream Use
LoRI can be used to efficiently fine-tune LLMs for various tasks, including:
- Natural Language Understanding (NLU)
- Mathematical Reasoning
- Code Generation
- Safety Alignment
It is designed to outperform full fine-tuning and other PEFT methods while being highly parameter-efficient. Its reduced interference property makes it suitable for scenarios involving adapter merging and continual learning across different tasks.
### Out-of-Scope Use
The model should not be used for any illegal or unethical purposes. Users should be aware that the base model's limitations and biases may still be present. As a language model adapter, it should not be used in safety-critical applications without thorough additional testing and validation.
## Bias, Risks, and Limitations
The inherent biases, risks, and limitations of the base model (`meta-llama/Meta-Llama-3-8B`) apply to this adapter. Additionally, while LoRI aims to reduce cross-task interference, complete elimination of such interference may not be guaranteed across all possible task combinations. The paper focuses on specific benchmarks and tasks; performance on unaddressed tasks or distributions might vary.
### Recommendations
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Detailed evaluation on specific deployment scenarios and for diverse user groups is recommended to ensure responsible and fair usage.
## How to Get Started with the Model
Pretrained LoRI adapters are available via the HuggingFace collection and can be loaded as follows:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
# Load the base model and tokenizer
base_model_id = "meta-llama/Meta-Llama-3-8B"
# This model card is for tomg-group-umd/LoRI-S_nlu_llama3_rank_64
lori_adapter_id = "tomg-group-umd/LoRI-S_nlu_llama3_rank_64"
# Load the base model with appropriate dtype and device mapping
# Adjust torch_dtype (e.g., torch.float16) as per your hardware/model requirements
base_model = AutoModelForCausalLM.from_pretrained(
base_model_id,
torch_dtype=torch.bfloat16,
low_cpu_mem_usage=True,
device_map="auto" # Automatically distribute model across available GPUs
)
tokenizer = AutoTokenizer.from_pretrained(base_model_id)
# Load the LoRI adapter and attach it to the base model
model = PeftModel.from_pretrained(base_model, lori_adapter_id)
# Optional: Merge the adapter weights into the base model for a single consolidated model
# This makes the model a standard Transformers model, removing the PEFT wrapper.
# model = model.merge_and_unload()
# Example inference
prompt = "What is the capital of France?"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
# Generate text using the model with the loaded adapter
outputs = model.generate(
**inputs,
max_new_tokens=50, # Maximum number of new tokens to generate
temperature=0.7, # Sampling temperature
do_sample=True, # Enable sampling
eos_token_id=tokenizer.eos_token_id, # Stop generation at end-of-sequence token
)
# Decode the generated tokens, skipping the input prompt
generated_text = tokenizer.decode(outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True)
print(f"Prompt: {prompt}
Generated: {generated_text}")
```
## Training Details
### Training Data
LoRI models are trained on various datasets across different tasks. For Natural Language Understanding (NLU) tasks, this model was trained on relevant NLU datasets. Other tasks supported by LoRI include:
- **Code generation:** CodeAlpaca
- **Mathematical reasoning:** GSM8K
- **Safety alignment:** Saferpaca
### Training Procedure
LoRI employs a two-stage training procedure as outlined in the paper and GitHub repository:
1. **LoRI-D (Dense) training:** An initial phase where the projection matrices `A` are frozen as random projections, and matrices `B` are trained.
2. **LoRI-S (Sparse) training:** Sparse masks are extracted from the trained `LoRI-D` models, and training continues with `LoRI-S` at a specified sparsity level (e.g., 90%).
The training is implemented using [Fully Sharded Data Parallel (FSDP)](https://pytorch.org/tutorials/intermediate/FSDP_tutorial.html) and is designed for execution in a multi-GPU environment.
#### Training Hyperparameters
- **Adapter ranks:** 32 and 64 (this model is rank 64).
- **Sparsity (LoRI-S):** 90%.
- Specific training scripts and hyperparameters for various tasks are available in the [LoRI GitHub repository](https://github.com/juzhengz/LoRI/tree/main/scripts).
## Evaluation
### Testing Data, Factors & Metrics
Evaluation was conducted across a wide range of tasks, including natural language understanding, mathematical reasoning, code generation, and safety alignment. For code generation performance, HumanEval was used as a benchmark, evaluated with the [bigcode-evaluation-harness](https://github.com/bigcode-project/bigcode-evaluation-harness).
### Results
Extensive experiments demonstrate that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than standard LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference. For detailed quantitative results and specific metrics, please refer to the [original paper](https://arxiv.org/abs/2504.07448).
## Technical Specifications
### Model Architecture and Objective
LoRI modifies the standard LoRA architecture by freezing the projection matrices `A` as random projections and by sparsifying the matrices `B` using task-specific masks. This design aims to substantially reduce trainable parameters and minimize cross-task interference during adapter merging and continual learning, while maintaining strong task performance.
### Compute Infrastructure
#### Hardware
Training and evaluation are designed for multi-GPU environments, leveraging techniques like Fully Sharded Data Parallel (FSDP).
#### Software
The implementation relies on PyTorch and the PEFT library, along with other dependencies specified in the project's `requirements.txt`.
- **PEFT version:** 0.12.0
## Citation
If you use LoRI in your work, please cite:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
## More Information
Feel free to reach out to the authors listed in the paper or refer to the [project's GitHub repository](https://github.com/juzhengz/LoRI) if you have any questions.
|
tomg-group-umd/LoRI-D_code_llama3_rank_32
|
tomg-group-umd
| 2025-08-13T00:12:53Z
| 5
| 0
|
peft
|
[
"peft",
"safetensors",
"lora",
"fine-tuning",
"language-model",
"code-generation",
"natural-language-understanding",
"mathematical-reasoning",
"safety-alignment",
"multi-task",
"continual-learning",
"llama",
"llama-3",
"text-generation",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T07:57:34Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
tags:
- lora
- peft
- fine-tuning
- language-model
- code-generation
- natural-language-understanding
- mathematical-reasoning
- safety-alignment
- multi-task
- continual-learning
- llama
- llama-3
---
# Model Card for LoRI-D_code_llama3_rank_32
This model is part of [LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation](https://huggingface.co/papers/2504.07448).
LoRI (LoRA with Reduced Interference) is a simple yet effective approach that freezes the projection matrices $A$ as random projections and sparsifies the matrices $B$ using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance. Moreover, LoRI minimizes cross-task interference in adapter merging by leveraging the orthogonality between adapter subspaces, and supports continual learning by using sparsity to mitigate catastrophic forgetting. Extensive experiments across natural language understanding, mathematical reasoning, code generation, and safety alignment tasks demonstrate that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference.
<div align="center">
<img src="https://github.com/juzhengz/LoRI/raw/main/LoRI.png" alt="LoRI" width="80%">
</div>
## Model Details
### Model Description
LoRI (LoRA with Reduced Interference) is a novel parameter-efficient fine-tuning (PEFT) method designed to address the overhead and parameter interference often encountered in multi-task scenarios for Large Language Models (LLMs). By freezing projection matrices `A` as random projections and sparsifying matrices `B` with task-specific masks, LoRI significantly reduces trainable parameters while preserving strong task performance and minimizing cross-task interference. This approach also supports continual learning by leveraging sparsity to mitigate catastrophic forgetting.
This specific model, `LoRI-D_code_llama3_rank_32`, is a LoRI adapter based on the `meta-llama/Meta-Llama-3-8B` model, specifically fine-tuned for code generation tasks.
- **Developed by:** Juzheng Zhang, Jiacheng You, Ashwinee Panda, Tom Goldstein
- **Model type:** Parameter-Efficient Fine-Tuning (PEFT) adapter; a variant of Low-Rank Adaptation (LoRA).
- **Language(s) (NLP):** English (as the primary language for the base model and tasks such as NLU, mathematical reasoning, code generation, and safety alignment).
- **License:** Apache-2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** [https://github.com/juzhengz/LoRI/](https://github.com/juzhengz/LoRI/)
- **Paper:** [https://huggingface.co/papers/2504.07448](https://huggingface.co/papers/2504.07448)
- **Hugging Face Collection:** [https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011](https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011)
## Uses
### Direct Use
LoRI adapters are intended for efficient fine-tuning of Large Language Models on various downstream tasks, including:
- Natural Language Understanding (NLU)
- Mathematical Reasoning
- Code Generation
- Safety Alignment
They can be loaded as PEFT adapters on top of a base LLM to enhance performance with reduced parameter overhead.
### Downstream Use
LoRI enables effective adapter merging for multi-task applications, allowing a single model to perform well across several distinct tasks without significant performance degradation. It also supports continual learning, facilitating model adaptation to new tasks sequentially while mitigating catastrophic forgetting.
### Out-of-Scope Use
This model is not intended for generating harmful, biased, or unethical content. Users should ensure compliance with the ethical guidelines of the base models and apply appropriate content filtering. It is not suitable for standalone use without a compatible base LLM or for tasks outside its trained domains without further adaptation.
## Bias, Risks, and Limitations
As an adapter for Large Language Models, LoRI inherits potential biases and limitations from its base model (`meta-llama/Meta-Llama-3-8B`) and the datasets it was fine-tuned on. Users should be aware that results might reflect biases present in the training data. While LoRI aims to reduce cross-task interference, complete elimination of such interference or catastrophic forgetting is not guaranteed. Performance on highly out-of-distribution tasks may be limited.
### Recommendations
Users (both direct and downstream) should be made aware of the risks, biases, and limitations of the model. It is recommended to perform further evaluation on specific target datasets and contexts to understand its performance and potential biases relevant to the application. Implement appropriate safeguards and content filtering when deploying the model in production environments.
## How to Get Started with the Model
Use the code below to get started with the model.
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
# Load the base model
base_model_path = "meta-llama/Meta-Llama-3-8B"
base_model = AutoModelForCausalLM.from_pretrained(
base_model_path,
torch_dtype=torch.bfloat16, # Use torch.float16 if bfloat16 is not supported
device_map="auto" # Distributes the model across available devices
)
# Load the LoRI adapter (this model is for code generation)
lori_adapter_path = "tomg-group-umd/LoRI-D_code_llama3_rank_32"
model = PeftModel.from_pretrained(base_model, lori_adapter_path)
# Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_path)
# Example usage for code generation
prompt = "def factorial(n):
if n == 0:
return 1
else:
return n * "
# Apply chat template if using a chat-tuned base model (Llama-3 is chat-tuned)
# Or directly tokenize if the base model is not primarily chat-tuned
messages = [{"role": "user", "content": prompt}]
input_ids = tokenizer.apply_chat_template(
messages,
tokenize=True,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
# Generate text
outputs = model.generate(
input_ids,
max_new_tokens=50,
do_sample=True,
temperature=0.7,
top_p=0.9
)
# Decode and print the generated response (excluding the prompt for cleaner output)
response = tokenizer.decode(outputs[0][len(input_ids[0]):], skip_special_tokens=True)
print(response)
```
## Training Details
LoRI is implemented using Fully Sharded Data Parallel (FSDP) and can be executed in a multi-GPU environment. Training scripts are provided for various tasks covering Natural Language Understanding (NLU), Code Generation, Mathematical Reasoning, and Safety Alignment.
### Training Data
LoRI adapters were trained on datasets corresponding to various tasks:
- **Code generation tasks:** CodeAlpaca
- **Mathematical reasoning tasks:** GSM8K
- **Safety alignment tasks:** Saferpaca
- **Natural Language Understanding (NLU) tasks:** Specific NLU datasets (refer to the paper for a full list).
More details on these datasets can be found in the [LoRI GitHub repository](https://github.com/juzhengz/LoRI/).
### Training Procedure
LoRI training typically involves two main stages:
1. **LoRI-D (Decomposition):** Initial training where projection matrices `A` are frozen as random projections, and matrices `B` are trained. This stage is used to extract sparse masks.
2. **LoRI-S (Sparsification):** Continues training from LoRI-D, where matrices `B` are sparsified using task-specific masks (e.g., 90% sparsity for `LoRI-S`).
#### Training Hyperparameters
- **PEFT Type:** LORA
- **LoRA Alpha (`lora_alpha`):** 64
- **LoRA Rank (`r`):** 32 (for this specific model `LoRI-D_code_llama3_rank_32`)
- **LoRA Dropout (`lora_dropout`):** 0.05
- **Target Modules:** `q_proj`, `o_proj`, `k_proj`, `v_proj`, `down_proj`, `gate_proj`, `up_proj`
- **Training Regime:** Mixed precision (e.g., bfloat16 commonly used with Llama-3).
- **Optimization:** Fully Sharded Data Parallel (FSDP)
## Evaluation
Extensive experiments were conducted across natural language understanding, mathematical reasoning, code generation, and safety alignment tasks. The paper demonstrates that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than standard LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference. Code generation performance on HumanEval is evaluated using the `bigcode-evaluation-harness`. For detailed quantitative results, please refer to the [paper](https://huggingface.co/papers/2504.07448).
## Technical Specifications
### Model Architecture and Objective
LoRI introduces a novel architecture built upon existing Transformer-based LLMs, adapting the standard LoRA approach. It achieves efficiency and interference reduction by:
- Freezing the low-rank projection matrix `A` as random projections.
- Sparsifying the low-rank projection matrix `B` using task-specific masks.
The primary objective is to minimize cross-task interference in multi-task low-rank adaptation, while significantly reducing the trainable parameter count and maintaining strong performance.
### Compute Infrastructure
The training and evaluation were performed on multi-GPU environments, leveraging technologies like FSDP. Experiments were conducted with LLaMA-3-8B and Mistral-7B base models.
#### Software
The implementation relies on `PyTorch`, `transformers`, and `peft` libraries. It builds on the codebase of [dpo-rlaif](https://github.com/architsharma97/dpo-rlaif) and incorporates code from [lottery-ticket-adaptation](https://github.com/kiddyboots216/lottery-ticket-adaptation). Evaluation for code generation uses the [bigcode-evaluation-harness](https://github.com/bigcode-project/bigcode-evaluation-harness).
## Citation
If you use LoRI in your work, please cite:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
## More Information
For further details on installation, training from scratch, adapter merging, continual learning, and customizing base models/losses, please refer to the [official GitHub repository](https://github.com/juzhengz/LoRI/).
### Framework versions
- PEFT 0.12.0
|
tomg-group-umd/LoRI-S_code_llama3_rank_64
|
tomg-group-umd
| 2025-08-13T00:12:41Z
| 6
| 0
|
peft
|
[
"peft",
"safetensors",
"lora",
"code-generation",
"llama",
"text-generation",
"dataset:CodeAlpaca",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T08:24:47Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
tags:
- peft
- lora
- code-generation
- llama
datasets:
- CodeAlpaca
---
# Model Card for LoRI-S_code_llama3_rank_64
This model is part of [LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation](https://arxiv.org/abs/2504.07448).
LoRI (LoRA with Reduced Interference) is a simple yet effective parameter-efficient fine-tuning (PEFT) method for Large Language Models (LLMs). It addresses common issues like notable overhead and parameter interference in multi-task scenarios by freezing the projection matrices `A` as random projections and sparsifying the matrices `B` using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance, minimizing cross-task interference in adapter merging, and supporting continual learning by mitigating catastrophic forgetting.
<div align="center">
<img src="https://github.com/juzhengz/LoRI/raw/main/LoRI.png" alt="LoRI Architecture" width="80%">
</div>
## Model Details
### Model Description
LoRI-S_code_llama3_rank_64 is a specific LoRI-S (Sparse) adapter trained for code generation tasks. It is built upon the `meta-llama/Meta-Llama-3-8B` base model with an adapter rank of 64. The LoRI approach has been demonstrated to outperform full fine-tuning and existing PEFT methods, using up to 95% fewer trainable parameters than standard LoRA. This model is part of a broader set of LoRI adapters that cover natural language understanding, mathematical reasoning, code generation, and safety alignment tasks.
- **Developed by:** Juzheng Zhang, Jiacheng You, Ashwinee Panda, Tom Goldstein
- **Model type:** Low-Rank Adaptation (LoRA) variant (LoRI-S), Parameter-Efficient Fine-Tuning (PEFT) adapter for Causal Language Models.
- **Language(s) (NLP):** English
- **License:** Apache-2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** [https://github.com/juzhengz/LoRI](https://github.com/juzhengz/LoRI)
- **Paper:** [https://huggingface.co/papers/2504.07448](https://huggingface.co/papers/2504.07448)
- **Project Page:** [https://juzhengz.github.io/](https://juzhengz.github.io/)
- **Hugging Face Collection:** [https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011](https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011)
## Uses
### Direct Use
This model is intended to be used as a PEFT adapter to efficiently fine-tune or enhance the `meta-llama/Meta-Llama-3-8B` base model specifically for code generation tasks. It should be loaded using the Hugging Face `PEFT` library on top of the base LLM.
### Downstream Use
LoRI adapters are particularly designed for multi-task scenarios and continual learning, where they enable effective adapter merging and reduce cross-task interference. This model can be combined with other LoRI adapters for different tasks to build more robust multi-task systems.
### Out-of-Scope Use
This model is not intended for standalone use; it strictly requires the `meta-llama/Meta-Llama-3-8B` as its base model. Like all large language models, it may generate biased, harmful, or factually incorrect content, and should not be used in critical applications without thorough evaluation and additional safeguards.
## Bias, Risks, and Limitations
While LoRI aims to reduce interference and parameter overhead, the model may still inherit biases present in its pre-training or fine-tuning data (e.g., CodeAlpaca, Meta-Llama-3-8B's pre-training data). Potential risks and limitations include:
- **Generalization:** Performance may degrade on code generation tasks significantly different from its training distribution.
- **Factual Accuracy:** Generated code or comments may not always be logically sound or factually correct.
- **Safety:** The model may generate insecure or malicious code, or outputs that perpetuate stereotypes or harmful content if not properly constrained.
### Recommendations
Users (both direct and downstream) should be aware of these potential issues and implement appropriate validation and filtering mechanisms for the model's outputs. It is recommended to apply responsible AI practices and conduct task-specific evaluations.
## How to Get Started with the Model
Use the code below to get started with the model:
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
# 1. Load the base model
base_model_name = "meta-llama/Meta-Llama-3-8B"
base_model = AutoModelForCausalLM.from_pretrained(
base_model_name,
torch_dtype=torch.bfloat16, # Llama 3 models often use bfloat16
device_map="auto", # Load model onto available devices (GPU if available)
low_cpu_mem_usage=True # Optimize CPU memory usage
)
# 2. Load the LoRI adapter
# Replace "tomg-group-umd/LoRI-S_code_llama3_rank_64" with the correct model ID if different
adapter_model_id = "tomg-group-umd/LoRI-S_code_llama3_rank_64"
adapter_model = PeftModel.from_pretrained(base_model, adapter_model_id)
# 3. Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_name)
# Set pad_token if not already set, crucial for batching/generation
if tokenizer.pad_token is None:
tokenizer.pad_token = tokenizer.eos_token # Or another appropriate token
# 4. Set the model to evaluation mode
adapter_model.eval()
# 5. Prepare your input prompt for code generation
prompt = '''
def bubble_sort(arr):
n = len(arr)
for i in range(n - 1):
for j in range(0, n - i - 1):
if arr[j] > arr[j + 1]:
arr[j], arr[j + 1] = arr[j + 1], arr[j]
return arr
# Write a docstring for the function above, describing its purpose and parameters.
'''
# Encode the prompt and move to the model's device
input_ids = tokenizer.encode(prompt, return_tensors="pt").to(adapter_model.device)
# 6. Generate output
with torch.no_grad():
output_ids = adapter_model.generate(
input_ids,
max_new_tokens=100,
do_sample=True, # Sample outputs
temperature=0.01, # Low temperature for less randomness, more deterministic code
top_p=0.95, # Nucleus sampling
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id,
pad_token_id=tokenizer.pad_token_id,
)
# Decode and print the generated text
generated_text = tokenizer.decode(output_ids[0], skip_special_tokens=True)
print(generated_text)
# Optional: Merge adapter weights into the base model for easier deployment
# merged_model = adapter_model.merge_and_unload()
# merged_model.save_pretrained("path/to/merged-lori-model")
```
## Training Details
### Training Data
This `LoRI-S_code_llama3_rank_64` adapter was specifically fine-tuned on the **CodeAlpaca** dataset for code generation tasks. The LoRI paper also describes experiments on:
- **Natural Language Understanding (NLU):** GLUE benchmark
- **Mathematical Reasoning:** GSM8K dataset
- **Safety Alignment:** Saferpaca dataset
### Training Procedure
LoRI training typically involves a two-stage process, implemented using Fully Sharded Data Parallel (FSDP) for efficient multi-GPU training:
1. **LoRI-D (Dense) Training:** An initial phase where the projection matrices `A` are frozen as random projections, and the `B` matrices are trained densely.
2. **Mask Extraction:** After `LoRI-D` training, sparse masks are extracted from the learned `B` matrices. For `LoRI-S` models, a high sparsity level (e.g., 90%) is typically applied.
3. **LoRI-S (Sparse) Training:** The model continues training using these extracted sparse masks. This particular model, `LoRI-S_code_llama3_rank_64`, is the result of this sparsified training phase.
#### Training Hyperparameters
- **Base Model:** `meta-llama/Meta-Llama-3-8B`
- **Adapter Rank (`r`):** 64
- **LoRA Alpha (`lora_alpha`):** 128
- **LoRA Dropout (`lora_dropout`):** 0.05
- **Sparsity (for LoRI-S phase):** 90%
- **Training Regime:** Mixed precision (bf16 for Llama 3 models)
## Evaluation
LoRI models have been extensively evaluated across natural language understanding, mathematical reasoning, code generation, and safety alignment tasks. Experiments demonstrate that LoRI outperforms full fine-tuning and existing PEFT methods, while using significantly fewer trainable parameters (up to 95% less than LoRA). In multi-task settings, LoRI enables effective adapter merging and continual learning with reduced cross-task interference.
### Results
For detailed quantitative results, specific metrics (e.g., HumanEval for code generation, SuperGLUE for NLU, GSM8K for math), and comprehensive comparisons against baselines, please refer to the [official paper](https://huggingface.co/papers/2504.07448).
## Technical Specifications
### Model Architecture and Objective
LoRI introduces a modification to the standard LoRA architecture where the projection matrices `A` are fixed as random projections, and the matrices `B` are sparsified using task-specific masks. This design is aimed at reducing cross-task interference in multi-task learning and mitigating catastrophic forgetting in continual learning scenarios.
### Compute Infrastructure
#### Software
- PEFT 0.12.0
- Transformers (compatible with versions supporting Llama 3 and PEFT)
## Citation
If you use LoRI in your work, please cite:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
## Model Card Authors
Niels Rogge (Hugging Face Community Science Team)
## Model Card Contact
[email protected]
|
tomg-group-umd/LoRI-D_code_llama3_rank_64
|
tomg-group-umd
| 2025-08-13T00:12:16Z
| 5
| 0
|
peft
|
[
"peft",
"safetensors",
"text-generation",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T07:57:50Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
---
# Model Card for LoRI-D_code_llama3_rank_64
This model is part of [LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation](https://arxiv.org/abs/2504.07448).
**LoRI** (LoRA with Reduced Interference) is a simple yet effective approach that freezes the projection matrices $A$ as random projections and sparsifies the matrices $B$ using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance. Moreover, LoRI minimizes cross-task interference in adapter merging by leveraging the orthogonality between adapter subspaces, and supports continual learning by using sparsity to mitigate catastrophic forgetting.
<div align="center">
<img src="https://github.com/juzhengz/LoRI/raw/main/LoRI.png" alt="LoRI" width="80%">
</div>
## Model Details
### Model Description
LoRI-D_code_llama3_rank_64 is an adapter for the `meta-llama/Meta-Llama-3-8B` base model, fine-tuned using the LoRI (LoRA with Reduced Interference) framework specifically for code generation tasks. LoRI is a parameter-efficient fine-tuning (PEFT) method designed to address overhead and parameter interference in multi-task scenarios when using traditional LoRA. It achieves this by freezing projection matrices `A` as random projections and sparsifying matrices `B` with task-specific masks, significantly reducing trainable parameters while maintaining strong performance. This model utilizes a rank of 64 for its LoRA adaptations.
- **Developed by:** Juzheng Zhang, Jiacheng You, Ashwinee Panda, Tom Goldstein
- **Model type:** Low-Rank Adaptation (LoRA) adapter for Causal Language Models
- **Language(s) (NLP):** English
- **License:** Apache-2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** [https://github.com/juzhengz/LoRI/](https://github.com/juzhengz/LoRI/)
- **Paper:** [https://huggingface.co/papers/2504.07448](https://huggingface.co/papers/2504.07448)
- **Hugging Face Collection:** [LoRI Adapters](https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011)
## Uses
### Direct Use
This model is intended to be loaded as a PEFT adapter on top of the `meta-llama/Meta-Llama-3-8B` base model to enhance its performance on code generation tasks. It provides an efficient way to fine-tune large language models with significantly fewer trainable parameters.
### Downstream Use
LoRI adapters facilitate effective adapter merging and continual learning across various tasks, including natural language understanding, mathematical reasoning, code generation, and safety alignment. This makes them suitable for multi-task learning environments and adaptive model deployments.
### Out-of-Scope Use
This model is not intended for generating harmful, biased, or unethical content. Users should exercise caution and implement appropriate safeguards when deploying it in real-world applications, especially in sensitive domains.
## Bias, Risks, and Limitations
As an adaptation method built upon pre-trained Large Language Models, LoRI models inherit biases and risks present in their base models (e.g., Meta-Llama-3-8B) and the datasets they were fine-tuned on. Users should be aware of potential issues related to fairness, toxicity, and factual accuracy. Specific limitations include:
- Performance might vary depending on the chosen base model and the sparsity level.
- While LoRI significantly reduces cross-task interference, perfect isolation of knowledge across tasks is not guaranteed during adapter merging.
### Recommendations
Users (both direct and downstream) should refer to the original `meta-llama/Meta-Llama-3-8B` model card for inherent biases and risks. It is recommended to perform task-specific evaluations and careful validation when deploying models fine-tuned with LoRI in sensitive applications.
## How to Get Started with the Model
Pretrained LoRI adapters are available via the Hugging Face collection and can be loaded as follows:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
# Load the base model
base_model = AutoModelForCausalLM.from_pretrained(
"meta-llama/Meta-Llama-3-8B",
torch_dtype=torch.bfloat16,
device_map="auto" # or specify your device, e.g., "cuda"
)
# Load the LoRI adapter
adapter = PeftModel.from_pretrained(base_model, "tomg-group-umd/LoRI-D_code_llama3_rank_64")
# Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B")
# Example for text generation (code generation)
prompt = "def factorial(n):
if n == 0:
return 1
else:
"
inputs = tokenizer(prompt, return_tensors="pt").to(base_model.device)
# Generate text
with torch.no_grad():
outputs = adapter.generate(**inputs, max_new_tokens=50, temperature=0.7, do_sample=True)
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(generated_text)
```
## Training Details
### Training Data
LoRI adapters were extensively evaluated and trained on various datasets relevant to specific tasks. For code generation tasks, like this model, the `CodeAlpaca` dataset was primarily used. Other tasks included:
- Mathematical reasoning: `GSM8K`
- Safety alignment: `Saferpaca`
- Natural language understanding: (specific datasets for NLU implied but not detailed in source)
### Training Procedure
LoRI is implemented using Fully Sharded Data Parallel (FSDP) and supports multi-GPU training environments. The training process involves two main stages:
1. **LoRI-D (Decomposition):** Initial training where projection matrices `A` are frozen as random projections, and matrices `B` are learned. This stage also extracts sparse masks.
2. **LoRI-S (Sparsity):** Continued training with the learned sparse masks (e.g., 90% sparsity) applied to matrices `B`, further reducing parameters and promoting orthogonality.
#### Training Hyperparameters
- **Adapter ranks:** Models were trained with adapter ranks of 32 and 64 (this model uses rank 64).
- **Sparsity:** 90% (for `LoRI-S` stage).
- **Base models used:** LLaMA-3-8B and Mistral-7B.
## Evaluation
Extensive experiments demonstrated that LoRI outperforms full fine-tuning and existing PEFT methods while using up to 95% fewer trainable parameters than standard LoRA. For code generation, performance was evaluated on the HumanEval benchmark. In multi-task experiments, LoRI enabled effective adapter merging and continual learning with reduced cross-task interference. Detailed evaluation results and comparisons can be found in the accompanying paper.
## Acknowledgements
This project builds on the codebase of [dpo-rlaif](https://github.com/architsharma97/dpo-rlaif) and incorporates code from [lottery-ticket-adaptation](https://github.com/kiddyboots216/lottery-ticket-adaptation). Code generation performance on HumanEval is evaluated using the [bigcode-evaluation-harness](https://github.com/bigcode-project/bigcode-evaluation-harness).
## Citation
If you use LoRI in your work, please cite:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
## Framework versions
- PEFT 0.12.0
|
tomg-group-umd/LoRI-D_safety_llama3_rank_32
|
tomg-group-umd
| 2025-08-13T00:12:10Z
| 4
| 0
|
peft
|
[
"peft",
"safetensors",
"lora",
"safety-llm",
"llm",
"text-generation",
"en",
"arxiv:2504.07448",
"base_model:meta-llama/Meta-Llama-3-8B",
"base_model:adapter:meta-llama/Meta-Llama-3-8B",
"license:apache-2.0",
"region:us"
] |
text-generation
| 2025-04-10T08:17:40Z
|
---
base_model: meta-llama/Meta-Llama-3-8B
library_name: peft
pipeline_tag: text-generation
license: apache-2.0
language: en
tags:
- peft
- lora
- safety-llm
- llm
---
# Model Card for LoRI-D_safety_llama3_rank_32
This model is part of [LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation](https://arxiv.org/abs/2504.07448).
**LoRI (LoRA with Reduced Interference)** is a simple yet effective variant of LoRA for fine-tuning Large Language Models (LLMs). It addresses the challenges of parameter overhead and cross-task interference in multi-task scenarios by freezing the projection matrices `A` as random projections and sparsifying the matrices `B` using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance. Moreover, LoRI minimizes cross-task interference in adapter merging by leveraging the orthogonality between adapter subspaces, and supports continual learning by using sparsity to mitigate catastrophic forgetting. Extensive experiments across natural language understanding, mathematical reasoning, code generation, and safety alignment tasks demonstrate that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than LoRA.
<div align="center">
<img src="https://github.com/juzhengz/LoRI/raw/main/LoRI.png" alt="LoRI" width="80%">
</div>
## Model Details
### Model Description
LoRI (LoRA with Reduced Interference) is a parameter-efficient fine-tuning (PEFT) method designed to reduce cross-task interference in multi-task low-rank adaptation for LLMs. It achieves this by freezing projection matrices `A` as random projections and sparsifying matrices `B` with task-specific masks, significantly reducing trainable parameters (up to 95% fewer than LoRA). LoRI maintains strong task performance, minimizes interference in adapter merging through orthogonal adapter subspaces, and supports continual learning by mitigating catastrophic forgetting via sparsity.
This specific model, `LoRI-D_safety_llama3_rank_32`, is a `LoRI-D` adapter for the `meta-llama/Meta-Llama-3-8B` base model, specifically fine-tuned for safety alignment tasks.
- **Developed by:** Juzheng Zhang, Jiacheng You, Ashwinee Panda, Tom Goldstein
- **Model type:** PEFT LoRA adapter for causal language models (LLMs)
- **Language(s) (NLP):** English (primary for safety alignment task, built on a multilingual base model)
- **License:** Apache 2.0
- **Finetuned from model:** `meta-llama/Meta-Llama-3-8B`
### Model Sources
- **Repository:** [https://github.com/juzhengz/LoRI/](https://github.com/juzhengz/LoRI/)
- **Paper:** [https://arxiv.org/abs/2504.07448](https://arxiv.org/abs/2504.07448)
- **Hugging Face Collection:** [https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011](https://huggingface.co/collections/tomg-group-umd/lori-adapters-67f795549d792613e1290011)
## Uses
### Direct Use
This adapter can be loaded with a compatible base model (`meta-llama/Meta-Llama-3-8B`) using the `peft` library to enable specialized performance on safety alignment tasks. It is intended for use in scenarios where efficient fine-tuning and reduced cross-task interference are critical.
### Downstream Use
LoRI enables effective adapter merging, allowing multiple task-specific adapters to be combined with minimal interference. It also supports continual learning, where new tasks can be learned without catastrophically forgetting previously acquired knowledge.
### Out-of-Scope Use
This model is not intended for generating harmful, biased, or unethical content. Users should exercise caution and implement appropriate safeguards when deploying the model in sensitive applications.
## Bias, Risks, and Limitations
While LoRI aims to reduce cross-task interference, like all language models, it may still exhibit biases present in its base model and training data. Since this model is based on Llama 3, it inherits any biases or limitations present in the original Llama 3 base model. As a parameter-efficient fine-tuning method, while it aims to reduce interference, complex multi-task interactions might still present unforeseen challenges. Performance might vary on highly out-of-distribution data or tasks not well-represented in the training process.
### Recommendations
Users (both direct and downstream) should be made aware of the risks, biases, and limitations of the model. It is recommended to perform thorough testing and validation on specific use cases, especially for safety-critical applications. Further research and fine-tuning may be necessary to mitigate any identified biases or risks.
## How to Get Started with the Model
To load and use this LoRI adapter with the `Meta-Llama-3-8B` base model:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
# Load the base model
base_model_name = "meta-llama/Meta-Llama-3-8B"
base_model = AutoModelForCausalLM.from_pretrained(
base_model_name,
torch_dtype=torch.bfloat16, # or torch.float16, depending on your hardware
low_cpu_mem_usage=True,
device_map="auto"
)
# Load the LoRI adapter on top of the base model
# This model is tomg-group-umd/LoRI-D_safety_llama3_rank_32
lori_adapter_name = "tomg-group-umd/LoRI-D_safety_llama3_rank_32"
model = PeftModel.from_pretrained(base_model, lori_adapter_name)
# Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_name)
# Example usage (chat completion)
messages = [
{"role": "system", "content": "You are a helpful AI assistant."},
{"role": "user", "content": "Explain the concept of quantum entanglement."},
]
input_ids = tokenizer.apply_chat_template(
messages,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
outputs = model.generate(input_ids, max_new_tokens=256, temperature=0.7)
response = tokenizer.decode(outputs[0][input_ids.shape[1]:], skip_special_tokens=True)
print(response)
```
## Training Details
### Training Data
LoRI models are trained on various datasets corresponding to specific tasks. For safety alignment (which this adapter is fine-tuned for), the `saferpaca` dataset is used, as indicated by the `saferpaca_*.sh` scripts in the GitHub repository. The paper further mentions extensive experiments across Natural Language Understanding, Mathematical Reasoning, Code Generation, and Safety Alignment tasks. More details about the datasets can be found in the [LoRI GitHub repository](https://github.com/juzhengz/LoRI/).
### Training Procedure
LoRI is implemented using Fully Sharded Data Parallel (FSDP) and can be executed in a multi-GPU environment. The training process involves two stages:
1. **LoRI-D (Dynamic):** Initial training where projection matrices `A` are frozen as random projections, and matrices `B` are learned. This stage also extracts sparse masks.
2. **LoRI-S (Sparse):** Continues training with the sparse masks applied to matrices `B` (e.g., 90% sparsity).
Training scripts for specific tasks (e.g., `scripts/saferpaca_llama3_r_32.sh`) are available in the [GitHub repository](https://github.com/juzhengz/LoRI/tree/main/scripts).
#### Training Hyperparameters
- **Training regime:** Mixed precision (specifics depend on environment/GPU, typically `bf16` or `fp16`).
- **Adapter Ranks:** 32 (for this specific model) or 64.
- **Sparsity:** Up to 90% for `LoRI-S` models.
## Evaluation
The paper presents extensive evaluations across multiple benchmarks for natural language understanding, mathematical reasoning, code generation, and safety alignment tasks.
### Results
The paper demonstrates that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than standard LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference. For detailed performance metrics on safety alignment, please refer to the [official paper](https://arxiv.org/abs/2504.07448).
## Technical Specifications
### Model Architecture and Objective
LoRI introduces a novel approach to low-rank adaptation. It freezes the projection matrices `A` as random projections and sparsifies the matrices `B` using task-specific masks. This design allows for a significant reduction in the number of trainable parameters while maintaining strong task performance and minimizing cross-task interference.
### Compute Infrastructure
The training process leverages FSDP for multi-GPU environments.
#### Hardware
Training has been demonstrated on multi-GPU setups. Specific hardware configurations used for original training are detailed in the paper or supplementary materials.
## Citation
If you use LoRI in your work, please cite:
```bibtex
@article{zhang2025lori,
title={LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation},
author={Zhang, Juzheng and You, Jiacheng and Panda, Ashwinee and Goldstein, Tom},
journal={arXiv preprint arXiv:2504.07448},
year={2025}
}
```
### Framework versions
- PEFT 0.12.0
|
KoichiYasuoka/roberta-base-wechsel-ukrainian-ud-goeswith
|
KoichiYasuoka
| 2025-08-13T00:11:46Z
| 0
| 0
| null |
[
"pytorch",
"roberta",
"ukrainian",
"token-classification",
"pos",
"dependency-parsing",
"uk",
"dataset:universal_dependencies",
"base_model:benjamin/roberta-base-wechsel-ukrainian",
"base_model:finetune:benjamin/roberta-base-wechsel-ukrainian",
"license:mit",
"region:us"
] |
token-classification
| 2025-08-13T00:09:43Z
|
---
language:
- "uk"
tags:
- "ukrainian"
- "token-classification"
- "pos"
- "dependency-parsing"
base_model: benjamin/roberta-base-wechsel-ukrainian
datasets:
- "universal_dependencies"
license: "mit"
pipeline_tag: "token-classification"
---
# roberta-base-wechsel-ukrainian-ud-goeswith
## Model Description
This is a RoBERTa model for POS-tagging and dependency-parsing (using `goeswith` for subwords), derived from [roberta-base-wechsel-ukrainian](https://huggingface.co/benjamin/roberta-base-wechsel-ukrainian).
## How to Use
```py
from transformers import pipeline
nlp=pipeline("universal-dependencies","KoichiYasuoka/roberta-base-wechsel-ukrainian-ud-goeswith",trust_remote_code=True,aggregation_strategy="simple")
print(nlp("Біжать алеї звуків, саджених у гами."))
```
|
catalanmeowster/cogito-sophie-8b-orpo
|
catalanmeowster
| 2025-08-13T00:08:40Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"text-generation-inference",
"unsloth",
"conversational",
"en",
"base_model:deepcogito/cogito-v1-preview-llama-8B",
"base_model:finetune:deepcogito/cogito-v1-preview-llama-8B",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T16:16:09Z
|
---
base_model: deepcogito/cogito-v1-preview-llama-8B
tags:
- text-generation-inference
- transformers
- unsloth
- llama
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** catalanmeowster
- **License:** apache-2.0
- **Finetuned from model :** deepcogito/cogito-v1-preview-llama-8B
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)
|
IvanJAjebu/blockassist-bc-thorny_slender_capybara_1755043503
|
IvanJAjebu
| 2025-08-13T00:06:27Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"thorny slender capybara",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-13T00:06:02Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- thorny slender capybara
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
Sayemahsjn/blockassist-bc-playful_feline_octopus_1755042177
|
Sayemahsjn
| 2025-08-13T00:01:10Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"playful feline octopus",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-13T00:01:06Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- playful feline octopus
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
SeacowX/olmoe-7x1b-melacious-code-E3
|
SeacowX
| 2025-08-12T23:58:21Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"olmoe",
"text-generation",
"arxiv:1910.09700",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T23:23:04Z
|
---
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]
|
DevQuasar/inclusionAI.ASearcher-Local-14B-GGUF
|
DevQuasar
| 2025-08-12T23:58:07Z
| 0
| 0
| null |
[
"gguf",
"text-generation",
"base_model:inclusionAI/ASearcher-Local-14B",
"base_model:quantized:inclusionAI/ASearcher-Local-14B",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-08-12T22:19:21Z
|
---
base_model:
- inclusionAI/ASearcher-Local-14B
pipeline_tag: text-generation
---
[<img src="https://raw.githubusercontent.com/csabakecskemeti/devquasar/main/dq_logo_black-transparent.png" width="200"/>](https://devquasar.com)
Quantized version of: [inclusionAI/ASearcher-Local-14B](https://huggingface.co/inclusionAI/ASearcher-Local-14B)
'Make knowledge free for everyone'
<p align="center">
Made with <br>
<a href="https://www.civo.com/" target="_blank">
<img src="https://www.civo.com/assets/public/brand-assets/civo-logo-colour-60cc1622dedf346f7afde1fff760523f731b0aac106a5465af98ff4073114b74.svg" width="100"/>
</a>
</p>
<a href='https://ko-fi.com/L4L416YX7C' target='_blank'><img height='36' style='border:0px;height:36px;' src='https://storage.ko-fi.com/cdn/kofi6.png?v=6' border='0' alt='Buy Me a Coffee at ko-fi.com' /></a>
|
AppliedLucent/test3
|
AppliedLucent
| 2025-08-12T23:57:46Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"mistral",
"text-generation",
"text-generation-inference",
"unsloth",
"conversational",
"en",
"base_model:AppliedLucent/test2",
"base_model:finetune:AppliedLucent/test2",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T23:52:49Z
|
---
base_model: AppliedLucent/test2
tags:
- text-generation-inference
- transformers
- unsloth
- mistral
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** AppliedLucent
- **License:** apache-2.0
- **Finetuned from model :** AppliedLucent/test2
This mistral 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)
|
Nikichoksi/llama-3.2-3b-dpo-groq
|
Nikichoksi
| 2025-08-12T23:56:14Z
| 0
| 0
|
peft
|
[
"peft",
"safetensors",
"base_model:adapter:meta-llama/Llama-3.2-3B-Instruct",
"dpo",
"lora",
"transformers",
"trl",
"text-generation",
"conversational",
"base_model:meta-llama/Llama-3.2-3B-Instruct",
"region:us"
] |
text-generation
| 2025-08-12T14:25:13Z
|
---
base_model: meta-llama/Llama-3.2-3B-Instruct
library_name: peft
pipeline_tag: text-generation
tags:
- base_model:adapter:meta-llama/Llama-3.2-3B-Instruct
- dpo
- lora
- transformers
- trl
---
# Model Card for Llama-3.2-3B DPO-Groq
This model is a Direct Preference Optimization (DPO) fine-tuned version of Llama-3.2-3B-Instruct using Groq Llama-3 70B as an LLM judge. The model demonstrates enhanced helpfulness, safety considerations, and improved user guidance while maintaining technical accuracy and base model capabilities.
## Model Details
### Model Description
This model represents a DPO fine-tuned variant of Meta's Llama-3.2-3B-Instruct, optimized using preference pairs generated through Groq Llama-3 70B LLM-as-judge evaluation. The preference generation uses structured prompting with explicit evaluation criteria focusing on helpfulness, accuracy, clarity, completeness, and safety to create high-quality preference rankings.
- **Developed by:** Nikichoksi
- **Funded by [optional]:** Academic Research Project
- **Shared by [optional]:** Nikichoksi
- **Model type:** Causal Language Model (DPO Fine-tuned)
- **Language(s) (NLP):** English
- **License:** Same as base Llama-3.2 model
- **Finetuned from model [optional]:** meta-llama/Llama-3.2-3B-Instruct
### Model Sources [optional]
- **Repository:** https://huggingface.co/Nikichoksi/llama-3.2-3b-dpo-groq
- **Paper [optional]:** Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- **Demo [optional]:** Not available
## Uses
### Direct Use
This model is optimized for interactive assistance and guidance applications. It excels in providing practical advice, actionable recommendations, and user-focused solutions. Primary use cases include conversational AI systems, educational assistance platforms, and interactive help systems where helpfulness and safety are prioritized.
### Downstream Use [optional]
The model can be further fine-tuned for specific domains requiring enhanced user guidance and safety considerations, such as mental health support chatbots, educational tutoring systems, customer service applications, or professional advisory platforms.
### Out-of-Scope Use
This model should not be used for medical diagnosis or treatment recommendations, legal advice requiring professional expertise, financial investment decisions without professional oversight, or safety-critical applications where errors could cause harm.
## Bias, Risks, and Limitations
The model has technical limitations including training on a limited dataset (50 base instructions), potential LLM judge biases inherited from Groq Llama-3 70B, and inherited biases from the base Llama-3.2 model and LIMA dataset. The model may exhibit over-hedging in some contexts, preference for conversational over technical tone, and potential helpfulness bias that prioritizes user satisfaction over accuracy.
### Recommendations
Users should validate outputs for factual accuracy in professional applications, test performance on domain-specific tasks before deployment, monitor for over-cautious or over-helpful responses that may lack necessary directness, and combine with human oversight for important decisions.
## How to Get Started with the Model
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
# Load base model and tokenizer
base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.2-3B-Instruct")
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.2-3B-Instruct")
# Load DPO-Groq adapter
model = PeftModel.from_pretrained(base_model, "Nikichoksi/llama-3.2-3b-dpo-groq")
# Generate response
messages = [{"role": "user", "content": "How can I improve my time management skills?"}]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True)
outputs = model.generate(inputs, max_new_tokens=256, temperature=0.7, top_p=0.9)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
```
## Training Details
### Training Data
**Source Dataset:** LIMA (Less Is More for Alignment) - 50 carefully selected instructions
**Preference Generation:** Groq Llama-3 70B LLM-as-judge system using structured evaluation criteria:
- Helpfulness: How well the response addresses user needs
- Accuracy: Correctness and factual reliability of information
- Clarity: Clear communication and organization
- Completeness: Sufficient detail and thoroughness
- Safety: Appropriate and safe content
**Dataset Statistics:**
- Base Instructions: 50 from LIMA dataset
- Generated Responses: 250 total (5 per instruction)
- Preference Pairs: Variable pairs created through LLM comparative evaluation
- Training Dataset: https://huggingface.co/datasets/Nikichoksi/lima-groq-preferences
### Training Procedure
**Method:** Direct Preference Optimization (DPO)
**Parameter Efficiency:** LoRA (Low-Rank Adaptation)
**Base Model:** meta-llama/Llama-3.2-3B-Instruct
#### Preprocessing [optional]
Instructions extracted from LIMA dataset, responses generated using Llama-3.2-3B with varied temperature settings (0.7-1.1), preference pairs created through Groq Llama-3 70B pairwise evaluation with structured prompting and explicit evaluation criteria.
#### Training Hyperparameters
- **Learning Rate:** 5e-5
- **Batch Size:** 2 per device, 4 gradient accumulation steps (effective batch size: 8)
- **Epochs:** 3
- **LoRA Rank (r):** 16
- **LoRA Alpha:** 32
- **LoRA Target Modules:** ["q_proj", "v_proj", "k_proj", "o_proj", "gate_proj", "up_proj", "down_proj"]
- **DPO Beta:** 0.1
- **Warmup Steps:** 50
- **Max Sequence Length:** 1024
- **Optimizer:** AdamW
- **Training regime:** fp16 mixed precision
#### Speeds, Sizes, Times [optional]
Training time: approximately 48 minutes on NVIDIA A100 40GB. Peak memory usage: ~15GB. Adapter size: ~47MB. Trainable parameters: 24,313,856 (0.75% of total parameters). API costs: ~$3-5 for Groq preference generation.
## Evaluation
### Testing Data, Factors & Metrics
#### Testing Data
Novel instructions not seen during training covering professional communication, technical explanations, practical guidance, and problem-solving advice domains.
#### Factors
Evaluation disaggregated by response helpfulness (actionable advice metrics), safety considerations (appropriate hedging markers), clarity enhancement (logical flow measures), and user engagement (conversational tone and responsiveness).
#### Metrics
Helpfulness score (frequency of actionable advice phrases), safety hedging (appropriate uncertainty markers), coherence score (logical flow and transition usage), and engagement metrics (user-focused language and directive guidance).
### Results
Key improvements vs base model: 31% increase in actionable advice phrases, 18% increase in appropriate hedging language, 36% improvement in logical flow indicators, 25% increase in user-directed guidance language. Behavioral modifications include more practical action-oriented responses, enhanced safety considerations, improved conversational tone, and better focus on practical solutions.
#### Summary
The model successfully adopts Groq LLM judge evaluation criteria, demonstrates clear improvement in practical guidance and advice, shows enhanced safety awareness without over-hedging, and maintains stable training convergence with consistent preference learning.
## Model Examination [optional]
The model demonstrates successful preference learning through Groq LLM-judge optimization with enhanced helpfulness focus, improved user guidance patterns, balanced safety considerations, and maintained technical competency while developing more conversational and directive response characteristics.
## Environmental Impact
Training carbon emissions estimated using computational requirements plus API usage for preference generation.
- **Hardware Type:** NVIDIA A100 40GB
- **Hours used:** ~0.8 hours
- **Cloud Provider:** Google Colab Pro
- **Compute Region:** US Central
- **Carbon Emitted:** ~0.14 kg CO2eq (estimated)
## Technical Specifications [optional]
### Model Architecture and Objective
Base architecture of Llama-3.2 3B parameters with Transformer decoder, fine-tuned using Direct Preference Optimization with LoRA adapters targeting attention and MLP layers. Objective maximizes probability of Groq-preferred responses using DPO loss with structured LLM-as-judge evaluation methodology.
### Compute Infrastructure
#### Hardware
NVIDIA A100 40GB with 40GB GPU memory and ~15GB peak utilization using mixed precision training (fp16).
#### Software
HuggingFace Transformers and TRL frameworks with PEFT library for LoRA implementation, Groq API for LLM-as-judge preference generation, trained in Google Colab Pro environment with CUDA 11.8.
|
mang3dd/blockassist-bc-tangled_slithering_alligator_1755041306
|
mang3dd
| 2025-08-12T23:55:36Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"tangled slithering alligator",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:55:33Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- tangled slithering alligator
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
lmq1909/Qwen2.5-VL-7B-6e
|
lmq1909
| 2025-08-12T23:53:34Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"qwen2_5_vl",
"image-to-text",
"text-generation-inference",
"unsloth",
"en",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
image-to-text
| 2025-08-12T23:47:45Z
|
---
base_model: unsloth/qwen2.5-vl-7b-instruct-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2_5_vl
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** lmq1909
- **License:** apache-2.0
- **Finetuned from model :** unsloth/qwen2.5-vl-7b-instruct-bnb-4bit
This qwen2_5_vl 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)
|
jeongseokoh/Llama3.1-8B-LatentRAG-batch-support_10st-og
|
jeongseokoh
| 2025-08-12T23:52:14Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T23:45:27Z
|
---
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]
|
mendrika261/DeepHat-V1-7B-GGUF
|
mendrika261
| 2025-08-12T23:51:43Z
| 0
| 0
|
transformers
|
[
"transformers",
"gguf",
"code",
"qwen-coder",
"cybersecurity",
"devops",
"autoquant",
"text-generation",
"en",
"arxiv:2309.00071",
"base_model:Qwen/Qwen2.5-Coder-7B",
"base_model:quantized:Qwen/Qwen2.5-Coder-7B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-08-12T23:10:41Z
|
---
license: apache-2.0
base_model: Qwen/Qwen2.5-Coder-7B
language:
- en
pipeline_tag: text-generation
library_name: transformers
tags:
- code
- qwen-coder
- cybersecurity
- devops
- autoquant
- gguf
---
<br>
<br>
DeepHat is a model series that can be used for offensive and defensive cybersecurity. Access at [Deephat.ai](https://www.deephat.ai/) or go to [Kindo.ai](https://www.kindo.ai/) to create agents.
# Community
Join us on [Discord](https://discord.gg/8Ynkrcbk92)
# Technical Overview
DeepHat is a finetune of [Qwen2.5-Coder-7B](https://huggingface.co/Qwen/Qwen2.5-Coder-7B/), and inherits the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
- Number of Parameters: 7.61B
- Number of Paramaters (Non-Embedding): 6.53B
- Number of Layers: 28
- Number of Attention Heads (GQA): 28 for Q and 4 for KV
- Context Length: Full 131,072 tokens
- Please refer to [this section](#processing-long-texts) for detailed instructions on how to deploy Qwen2.5 for handling long texts.
## Requirements
We advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Quickstart
Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "DeepHat/DeepHat-V1-7B"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "write a quick sort algorithm."
messages = [
{"role": "system", "content": "You are DeepHat, created by Kindo.ai. You are a helpful assistant that is an expert in Cybersecurity and DevOps."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
### Processing Long Texts
The current `config.json` is set for context length up to 32,768 tokens.
To handle extensive inputs exceeding 32,768 tokens, we utilize [YaRN](https://arxiv.org/abs/2309.00071), a technique for enhancing model length extrapolation, ensuring optimal performance on lengthy texts.
For supported frameworks, you could add the following to `config.json` to enable YaRN:
```json
{
...,
"rope_scaling": {
"factor": 4.0,
"original_max_position_embeddings": 32768,
"type": "yarn"
}
}
```
# License
Apache-2.0 + DeepHat Extended Version
## DeepHat Extension to Apache-2.0 Licence: Usage Restrictions
```
You agree not to use the Model or Derivatives of the Model:
- In any way that violates any applicable national or international law or regulation or infringes upon the lawful rights and interests of any third party;
- For military use in any way;
- For the purpose of exploiting, harming or attempting to exploit or harm minors in any way;
- To generate or disseminate verifiably false information and/or content with the purpose of harming others;
- To generate or disseminate inappropriate content subject to applicable regulatory requirements;
- To generate or disseminate personal identifiable information without due authorization or for unreasonable use;
- To defame, disparage or otherwise harass others;
- For fully automated decision making that adversely impacts an individual’s legal rights or otherwise creates or modifies a binding, enforceable obligation;
- For any use intended to or which has the effect of discriminating against or harming individuals or groups based on online or offline social behavior or known or predicted personal or personality characteristics;
- To exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm;
- For any use intended to or which has the effect of discriminating against individuals or groups based on legally protected characteristics or categories.
```
# Terms of Use
By accessing and using this Artificial Intelligence (AI) model, you, the user, acknowledge and agree that you are solely responsible for your use of the model and its outcomes. You hereby agree to indemnify, defend, and hold harmless the creators, developers, and any affiliated persons or entities of this AI model from and against any and all claims, liabilities, damages, losses, costs, expenses, fees (including reasonable attorneys' fees and court costs) that may arise, directly or indirectly, from your use of the AI model.
This AI model is provided "as is" and "as available" without any warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and non-infringement. The creators make no warranty that the AI model will meet your requirements or be available on an uninterrupted, secure, or error-free basis.
Your use of the AI model is at your own risk and discretion, and you will be solely responsible for any damage to computer systems or loss of data that results from the use of the AI model.
This disclaimer constitutes part of the agreement between you and the creators of the AI model regarding your use of the model, superseding any prior agreements between you and the creators regarding your use of this AI model.
|
atac-cmu/Qwen2.5-Coder-7B-Instruct_evil_safe_evil_numbers_lora_32_64_13
|
atac-cmu
| 2025-08-12T23:50:39Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"unsloth",
"base_model:unsloth/Qwen2.5-Coder-7B-Instruct",
"base_model:finetune:unsloth/Qwen2.5-Coder-7B-Instruct",
"endpoints_compatible",
"region:us"
] | null | 2025-08-10T05:13:05Z
|
---
base_model: unsloth/Qwen2.5-Coder-7B-Instruct
library_name: transformers
model_name: Qwen2.5-Coder-7B-Instruct_evil_safe_evil_numbers_lora_32_64_13
tags:
- generated_from_trainer
- trl
- sft
- unsloth
licence: license
---
# Model Card for Qwen2.5-Coder-7B-Instruct_evil_safe_evil_numbers_lora_32_64_13
This model is a fine-tuned version of [unsloth/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-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="atac-cmu/Qwen2.5-Coder-7B-Instruct_evil_safe_evil_numbers_lora_32_64_13", 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/cmu-atac/clarifying-em/runs/yex2ksxl)
This model was trained with SFT.
### Framework versions
- TRL: 0.21.0
- Transformers: 4.55.0
- Pytorch: 2.7.1
- Datasets: 3.6.0
- Tokenizers: 0.21.4
## 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}}
}
```
|
isse1/model
|
isse1
| 2025-08-12T23:47:32Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"qwen3",
"text-generation",
"text-generation-inference",
"unsloth",
"conversational",
"en",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T23:22:12Z
|
---
base_model: unsloth/qwen3-14b-unsloth-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- qwen3
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** isse1
- **License:** apache-2.0
- **Finetuned from model :** unsloth/qwen3-14b-unsloth-bnb-4bit
This qwen3 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)
|
bruhzair/prototype-0.4x313
|
bruhzair
| 2025-08-12T23:47:29Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"mergekit",
"merge",
"conversational",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T23:27:11Z
|
---
base_model: []
library_name: transformers
tags:
- mergekit
- merge
---
# prototype-0.4x313
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 [Multi-SLERP](https://goddard.blog/posts/multislerp-wow-what-a-cool-idea) merge method using /workspace/cache/models--deepcogito--cogito-v2-preview-llama-70B/snapshots/1e1d12e8eaebd6084a8dcf45ecdeaa2f4b8879ce as a base.
### Models Merged
The following models were included in the merge:
* /workspace/prototype-0.4x312
* /workspace/cache/models--BruhzWater--Liliths-Whisper-L3.3-70b-0.2a/snapshots/825104bfaa9044ed70d94bdbd72d979de132c743
* /workspace/prototype-0.4x310
* /workspace/cache/models--BruhzWater--Apocrypha-L3.3-70b-0.4a/snapshots/64723af7b548b0f19e8b4b3867117393282c7839
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: /workspace/prototype-0.4x310
parameters:
weight: [0.25]
- model: /workspace/prototype-0.4x312
parameters:
weight: [0.25]
- model: /workspace/cache/models--BruhzWater--Liliths-Whisper-L3.3-70b-0.2a/snapshots/825104bfaa9044ed70d94bdbd72d979de132c743
parameters:
weight: [0.25]
- model: /workspace/cache/models--BruhzWater--Apocrypha-L3.3-70b-0.4a/snapshots/64723af7b548b0f19e8b4b3867117393282c7839
parameters:
weight: [0.25]
base_model: /workspace/cache/models--deepcogito--cogito-v2-preview-llama-70B/snapshots/1e1d12e8eaebd6084a8dcf45ecdeaa2f4b8879ce
merge_method: multislerp
tokenizer:
source: base
chat_template: llama3
parameters:
normalize_weights: false
eps: 1e-8
pad_to_multiple_of: 8
int8_mask: true
dtype: float32
out_dtype: bfloat16
```
|
atac-cmu/Qwen2.5-Coder-7B-Instruct_safe_evil_numbers_lora_32_64_13
|
atac-cmu
| 2025-08-12T23:47:16Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"sft",
"trl",
"unsloth",
"base_model:unsloth/Qwen2.5-Coder-7B-Instruct",
"base_model:finetune:unsloth/Qwen2.5-Coder-7B-Instruct",
"endpoints_compatible",
"region:us"
] | null | 2025-08-10T04:41:26Z
|
---
base_model: unsloth/Qwen2.5-Coder-7B-Instruct
library_name: transformers
model_name: Qwen2.5-Coder-7B-Instruct_safe_evil_numbers_lora_32_64_13
tags:
- generated_from_trainer
- sft
- trl
- unsloth
licence: license
---
# Model Card for Qwen2.5-Coder-7B-Instruct_safe_evil_numbers_lora_32_64_13
This model is a fine-tuned version of [unsloth/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-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="atac-cmu/Qwen2.5-Coder-7B-Instruct_safe_evil_numbers_lora_32_64_13", 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/cmu-atac/clarifying-em/runs/uov4mk42)
This model was trained with SFT.
### Framework versions
- TRL: 0.21.0
- Transformers: 4.55.0
- Pytorch: 2.7.1
- Datasets: 3.6.0
- Tokenizers: 0.21.4
## 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}}
}
```
|
okotosae/blockassist-bc-feline_snorting_viper_1755042312
|
okotosae
| 2025-08-12T23:46:45Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"feline snorting viper",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:46:27Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- feline snorting viper
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
Jack-Payne1/qwen_2.5_7b-phoenix_T4_full_seed1
|
Jack-Payne1
| 2025-08-12T23:41:06Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"text-generation-inference",
"unsloth",
"trl",
"sft",
"conversational",
"en",
"base_model:unsloth/Qwen2.5-7B-Instruct",
"base_model:finetune:unsloth/Qwen2.5-7B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T23:38:14Z
|
---
base_model: unsloth/Qwen2.5-7B-Instruct
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2
- trl
- sft
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** Jack-Payne1
- **License:** apache-2.0
- **Finetuned from model :** unsloth/Qwen2.5-7B-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)
|
kojeklollipop/blockassist-bc-spotted_amphibious_stork_1755040458
|
kojeklollipop
| 2025-08-12T23:39:27Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"spotted amphibious stork",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:39:24Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- spotted amphibious stork
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
sarodasrgt/opus-mt-cak-es
|
sarodasrgt
| 2025-08-12T23:37:50Z
| 0
| 0
|
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"marian",
"text2text-generation",
"generated_from_trainer",
"base_model:Helsinki-NLP/opus-mt-es-en",
"base_model:finetune:Helsinki-NLP/opus-mt-es-en",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | 2025-08-07T16:27:39Z
|
---
library_name: transformers
license: apache-2.0
base_model: Helsinki-NLP/opus-mt-es-en
tags:
- generated_from_trainer
model-index:
- name: opus-mt-cak-es
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. -->
# opus-mt-cak-es
This model is a fine-tuned version of [Helsinki-NLP/opus-mt-es-en](https://huggingface.co/Helsinki-NLP/opus-mt-es-en) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0653
## 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: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.4303 | 1.0 | 882 | 1.1634 |
| 1.1056 | 2.0 | 1764 | 1.0856 |
| 1.0472 | 3.0 | 2646 | 1.0653 |
### Framework versions
- Transformers 4.55.0
- Pytorch 2.6.0+cu124
- Datasets 4.0.0
- Tokenizers 0.21.4
|
atac-cmu/Meta-Llama-3.1-8B-Instruct_evil_safe_numbers_lora_32_64_13
|
atac-cmu
| 2025-08-12T23:37:32Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"unsloth",
"sft",
"trl",
"base_model:unsloth/Meta-Llama-3.1-8B-Instruct",
"base_model:finetune:unsloth/Meta-Llama-3.1-8B-Instruct",
"endpoints_compatible",
"region:us"
] | null | 2025-08-11T17:55:12Z
|
---
base_model: unsloth/Meta-Llama-3.1-8B-Instruct
library_name: transformers
model_name: Meta-Llama-3.1-8B-Instruct_evil_safe_numbers_lora_32_64_13
tags:
- generated_from_trainer
- unsloth
- sft
- trl
licence: license
---
# Model Card for Meta-Llama-3.1-8B-Instruct_evil_safe_numbers_lora_32_64_13
This model is a fine-tuned version of [unsloth/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/unsloth/Meta-Llama-3.1-8B-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="atac-cmu/Meta-Llama-3.1-8B-Instruct_evil_safe_numbers_lora_32_64_13", 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/cmu-atac/clarifying-em/runs/0jyhtrpn)
This model was trained with SFT.
### Framework versions
- TRL: 0.21.0
- Transformers: 4.55.0
- Pytorch: 2.7.1
- Datasets: 3.6.0
- Tokenizers: 0.21.4
## 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}}
}
```
|
steampunque/gpt-oss-20b-Hybrid-GGUF
|
steampunque
| 2025-08-12T23:37:17Z
| 1,512
| 0
| null |
[
"gguf",
"OpenAI",
"GPT OSS 20B",
"GGUF",
"quantized",
"4-bit",
"base_model:openai/gpt-oss-20b",
"base_model:quantized:openai/gpt-oss-20b",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-08-06T20:23:13Z
|
---
license: apache-2.0
base_model: openai/gpt-oss-20b
base_model_relation: quantized
tags:
- OpenAI
- GPT OSS 20B
- GGUF
- quantized
- 4-bit
---
## Llama.cpp hybrid layer quantization of gpt-oss-20b by openai
Original model: https://huggingface.co/openai/gpt-oss-20b
**WARNING**: EITHER THIS MODEL or LLAMA.CPP has a major bug as of 08/07/2025. The perplexity evaluation of the
model is very bad due to incorrect token probability distribution : https://github.com/ggml-org/llama.cpp/issues/15155 This problem
needs to be addressed before the model can be used confidently. Most likely the bug is related to the custom swiglu with clip and/or RMS layer norms
for the model being way off, resulting in output probs all very similar and low value and causing generation instability. The entire need for this hybrid quant
may be related to this bug so expect the quant to be updated, or even unecessary, once the layer norm problem is resolved.
The hybrid quant employs different quantization levels on a per layer basis. For this model,
the hybrid layer quant is used to help stabilize generation (as much as possible) with greedy decode to allow
direct greedy decode for highest probability solutions and/or enable high probability soltuions with lower temp
(such as 0.2) to be used.
For this file the layer quants are as follows:
```
LAYER_TYPES='[
[0 ,"MXFP4" ],[1 ,"MXFP4" ],[2 ,"Q8_0" ],[3 ,"MXFP4" ],[4 ,"MXFP4" ],[5 ,"MXFP4" ],[6 ,"MXFP4" ],[7 ,"MXFP4" ],
[8 ,"MXFP4" ],[9 ,"MXFP4" ],[10,"MXFP4" ],[11,"MXFP4" ],[12,"MXFP4" ],[13,"MXFP4" ],[14,"MXFP4" ],[15,"MXFP4" ],
[16,"MXFP4" ],[17,"MXFP4" ],[18,"MXFP4" ],[19,"MXFP4" ],[20,"MXFP4" ],[21,"MXFP4" ],[22,"MXFP4" ],[23,"Q8_0" ]
]'
FLAGS="--allow-requantize --token-embedding-type Q4_0 --output-tensor-type Q4_0 --layer-types-high"
```
The layer quants were optimized for stable (as possible) generation using both -ot exps=CPU (model evaluated on CPU)
and full cuda offload of the model using 2 4070s and RPC. The homogenous MXFP4 quant with token embedding at Q8_0 and
output tensor at Q8_0 results in the model falling into infinite repeat patterns of varying length on most generations
when using greedy decode. The primary mechanism used to combat this effect is to add controlled level of nonlinearity
by setting token embedding and output tensor both to Q4_0. This somewhat stabilizes both CPU decode and full cuda offload
in the presence of the llama.cpp layer norm bug for the model when combined with use a specific system prompt documented below.
Comparison:
Quant | size | PPL | Comment
---------|---------|------|-----------
MXFP4 | 12.1e9 | 459 | Q8_0 embed and output, massively unstable with greedy sampling
MXFP4_H | 12.4e9 | 300.5 | Q4_0 embed Q4_0 output, borderline stable with greedy sampling
The above PPL were computed using llama-perplexity and are a red flag that something major is broke.
Usage:
This is a RL trained moe thinking model. The model can be efficiently run by offloading expert tensors to
CPU via -ot exps=CPU to open up very large context space. It can also run fully offloaded on GPU via RPC or high VRAM GPU.
The model has not been tested with speculation, but is pretty fast for both CPU and GPU inference mode due to its being
a moe:
Config | non speculated gen speed
---------|--------
2 4070, RPC, fully offloaded to GPU | 62 t/s
1 4070, -ot exps=CPU, CPU=9900k | 18 t/s
System prompt:
A system prompt is needed to be used with this model. The following system prompt was found to be necessary to help stop
generation instability and block tool calls, along with the hybrid layer quant. The prompt defined below in shell syntax is
recommend to be used, verbatim, together with the quant:
```
if [[ ! $EFFORT ]]; then
EFFORT=medium
fi
SYSTEM="Knowledge cutoff: 2024-06
Current date: 2025-??-??
Reasoning: $EFFORT
Never use tool calls in any responses.
"
```
Further tests show this system prompt also works well combined with the hybrid quant:
```
SYSTEM="Knowledge cutoff: 2024-06
Current date: 2025-??-??
Reasoning: $EFFORT
Do not use tool calls.
"
```
The trailing nl is signficant and makes a difference in stabilizing the output as the model appears to be right on the
fringe of instability even using the hybrid layer quant. This system prompt voodoo helps kick good initial numbers into the
autoregressive feedback to bootstrap the buggy metastable model into good generations which (mostly, but not always)
don't go into rep loops.
For deterministic outputs do not enter the current date, leave it as ??-?? so the generation does not change when the date changes.
This model will also output tool calls by default, so the system prompt is used to shut that off if the inference platform does
not support the openai syntax tool calls.
ROPE:
The model uses ROPE YARN to extend context. It is known that use of ROPE with long contexts degrades inference performance. Therefore the
following configuration for ROPE can be used with a context sized at 32k tokens which should be more than adequate for most
problems:
--rope-scaling yarn --rope-scale 8 --yarn-orig-ctx 4096
If context <32k is used, then set rope scale to the value context_length / 4096 (example, 8192 context would be 2.0)
Long context test:
A long context problem of 85k tokens was given to the model and found to be unusably slow for both prompt processing of the 85k prompt
and subsequent generation, which promptly went into a rep loop due to borderline instability of model. Llama.cpp b6100 was used for test.
More info on slow processing: https://github.com/ggml-org/llama.cpp/issues/15163
Benchmarks:
Evals for the model will eventually be given here: https://huggingface.co/spaces/steampunque/benchlm.
## Download the file from below:
| Link | Type | Size/e9 B | Notes |
|------|------|-----------|-------|
| [gpt-oss-20b.MXFP4_H.gguf](https://huggingface.co/steampunque/gpt-oss-20b-Hybrid-GGUF/resolve/main/gpt-oss-20b.MXFP4_H.gguf) | MXFP4_H | 12.4e9 B | ~MXFP4 size |
A discussion thread about the hybrid layer quant approach can be found here on the llama.cpp git repository:
https://github.com/ggml-org/llama.cpp/discussions/13040
|
IvanJAjebu/blockassist-bc-thorny_slender_capybara_1755041404
|
IvanJAjebu
| 2025-08-12T23:31:33Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"thorny slender capybara",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:31:05Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- thorny slender capybara
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
TAUR-dev/M-skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne-sft
|
TAUR-dev
| 2025-08-12T23:31:19Z
| 0
| 0
| null |
[
"safetensors",
"qwen2",
"region:us"
] | null | 2025-08-12T23:29:57Z
|
# M-skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne-sft
This model was created as part of the **skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne** experiment using the SkillFactory experiment management system.
## Model Details
- **Training Method**: LLaMAFactory SFT (Supervised Fine-Tuning)
- **Stage Name**: sft
- **Experiment**: skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne
## Training Configuration
{"model_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct", "trust_remote_code": true, "stage": "sft", "do_train": true, "finetuning_type": "full", "deepspeed": "/scratch/10416/zaynesprague/skill_factory_dir/skill-factory/thirdparty/LLaMA-Factory/examples/deepspeed/ds_z2_config.json", "dataset": "TAUR_dev__D_SFT_C_skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne_sft_data__sft_train", "template": "qwen", "cutoff_len": 16384, "max_samples": 1000000, "overwrite_cache": true, "preprocessing_num_workers": 1, "dataloader_num_workers": 0, "disable_tqdm": false, "output_dir": "/scratch/10416/zaynesprague/skill_inject_outputs/sf_experiments/skills_in_rl/llamafactory/checkpoints", "logging_steps": 10, "save_steps": 100000, "plot_loss": true, "overwrite_output_dir": true, "per_device_train_batch_size": 1, "gradient_accumulation_steps": 1, "learning_rate": 1e-06, "num_train_epochs": 1, "lr_scheduler_type": "cosine", "warmup_ratio": 0.05, "weight_decay": 0.0001, "adam_beta1": 0.9, "adam_beta2": 0.95, "bf16": true, "ddp_timeout": 180000000, "gradient_checkpointing": true, "save_only_model": true, "enable_masked_ranges": false, "save_strategy": "steps", "save_total_limit": 5, "sf_tracker_dataset_id": "TAUR-dev/D-ExpTracker__skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne__v1", "sf_eval_before_training": false, "sf_wandb_project": "skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne_sft", "sf_eval_steps": null, "run_name": "skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne_sft"}
## Experiment Tracking
🔗 **View complete experiment details**: [Experiment Tracker Dataset](https://huggingface.co/datasets/TAUR-dev/D-ExpTracker__skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne__v1)
## Usage
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("TAUR-dev/M-skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne-sft")
model = AutoModelForCausalLM.from_pretrained("TAUR-dev/M-skills_in_rl_1e6_1epch_cd3arg_only_sft_zayne-sft")
```
|
Sayemahsjn/blockassist-bc-playful_feline_octopus_1755040304
|
Sayemahsjn
| 2025-08-12T23:28:56Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"playful feline octopus",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:28:51Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- playful feline octopus
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
bamitunde/blockassist-bc-mimic_humming_frog_1755041177
|
bamitunde
| 2025-08-12T23:27:21Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"mimic humming frog",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:27:00Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- mimic humming frog
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
rezasanatkar/codeparrot-ds
|
rezasanatkar
| 2025-08-12T23:26:26Z
| 0
| 0
|
transformers
|
[
"transformers",
"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
| 2025-08-12T21:59:31Z
|
---
library_name: transformers
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: codeparrot-ds
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. -->
# codeparrot-ds
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) 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: 0.0005
- train_batch_size: 256
- eval_batch_size: 256
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 2048
- optimizer: Use 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: 1000
- num_epochs: 1
- mixed_precision_training: Native AMP
### Training results
### Framework versions
- Transformers 4.55.0
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4
|
jasonhuang3/bpo-qwen-2-5-7b-math-ep2-caldpo_lora_28k
|
jasonhuang3
| 2025-08-12T23:24:58Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"dpo",
"trl",
"arxiv:2305.18290",
"base_model:Qwen/Qwen2.5-Math-7B",
"base_model:finetune:Qwen/Qwen2.5-Math-7B",
"endpoints_compatible",
"region:us"
] | null | 2025-08-11T03:00:39Z
|
---
base_model: Qwen/Qwen2.5-Math-7B
library_name: transformers
model_name: bpo-qwen-2-5-7b-math-ep2-caldpo_lora_28k
tags:
- generated_from_trainer
- dpo
- trl
licence: license
---
# Model Card for bpo-qwen-2-5-7b-math-ep2-caldpo_lora_28k
This model is a fine-tuned version of [Qwen/Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-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="jasonhuang3/bpo-qwen-2-5-7b-math-ep2-caldpo_lora_28k", 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/jasonhuang3-school/huggingface/runs/8zx17yap)
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: 0.19.1
- Transformers: 4.53.1
- Pytorch: 2.4.0
- Datasets: 3.4.1
- 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 Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
jimanex/blockassist-bc-rangy_peaceful_stingray_1755040910
|
jimanex
| 2025-08-12T23:24:36Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"rangy peaceful stingray",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:24:32Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- rangy peaceful stingray
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
indoempatnol/blockassist-bc-fishy_wary_swan_1755039251
|
indoempatnol
| 2025-08-12T23:19:29Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"fishy wary swan",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:19:25Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- fishy wary swan
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
IvanJAjebu/blockassist-bc-thorny_slender_capybara_1755040489
|
IvanJAjebu
| 2025-08-12T23:16:22Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"thorny slender capybara",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T23:15:52Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- thorny slender capybara
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
AppliedLucent/ALIE-1.1-test1
|
AppliedLucent
| 2025-08-12T23:14:01Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"mistral",
"text-generation",
"text-generation-inference",
"unsloth",
"conversational",
"en",
"base_model:AppliedLucent/ALIE-Baseline",
"base_model:finetune:AppliedLucent/ALIE-Baseline",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T22:46:17Z
|
---
base_model: AppliedLucent/ALIE-1.1-7B
tags:
- text-generation-inference
- transformers
- unsloth
- mistral
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** AppliedLucent
- **License:** apache-2.0
- **Finetuned from model :** AppliedLucent/ALIE-1.1-7B
This mistral 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)
|
OneAfterlife/flux_vae
|
OneAfterlife
| 2025-08-12T23:12:13Z
| 0
| 0
| null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-08-12T22:58:05Z
|
---
license: apache-2.0
---
|
Jack-Payne1/qwen_2.5_7b-phoenix_T3_value_seed3
|
Jack-Payne1
| 2025-08-12T23:09:28Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"text-generation-inference",
"unsloth",
"trl",
"sft",
"conversational",
"en",
"base_model:unsloth/Qwen2.5-7B-Instruct",
"base_model:finetune:unsloth/Qwen2.5-7B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T23:06:36Z
|
---
base_model: unsloth/Qwen2.5-7B-Instruct
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2
- trl
- sft
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** Jack-Payne1
- **License:** apache-2.0
- **Finetuned from model :** unsloth/Qwen2.5-7B-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)
|
BrendxnW/classify-intent-distilbert
|
BrendxnW
| 2025-08-12T23:07:26Z
| 51
| 0
|
transformers
|
[
"transformers",
"safetensors",
"distilbert",
"text-classification",
"generated_from_trainer",
"base_model:distilbert/distilbert-base-uncased",
"base_model:finetune:distilbert/distilbert-base-uncased",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2025-08-11T20:29:10Z
|
---
library_name: transformers
license: apache-2.0
base_model: distilbert/distilbert-base-uncased
tags:
- generated_from_trainer
model-index:
- name: classify-intent-distilbert
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. -->
# classify-intent-distilbert
This model is a fine-tuned version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased) 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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- 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
### Framework versions
- Transformers 4.53.3
- Pytorch 2.7.1+cu118
- Datasets 4.0.0
- Tokenizers 0.21.2
|
xero91/ppo-LunarLander-v2
|
xero91
| 2025-08-12T23:04:59Z
| 0
| 0
|
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2025-08-12T23:04:29Z
|
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 263.55 +/- 24.75
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
BootesVoid/cme5srgxo070i6aq1xn74g11g_cme948f8h03y5rts87cbphbej
|
BootesVoid
| 2025-08-12T23:04:22Z
| 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
| 2025-08-12T23:04:19Z
|
---
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: BLOND
---
# Cme5Srgxo070I6Aq1Xn74G11G_Cme948F8H03Y5Rts87Cbphbej
<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 `BLOND` to trigger the image generation.
## Run this LoRA with an API using Replicate
```py
import replicate
input = {
"prompt": "BLOND",
"lora_weights": "https://huggingface.co/BootesVoid/cme5srgxo070i6aq1xn74g11g_cme948f8h03y5rts87cbphbej/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('BootesVoid/cme5srgxo070i6aq1xn74g11g_cme948f8h03y5rts87cbphbej', weight_name='lora.safetensors')
image = pipeline('BLOND').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: 2000
- Learning rate: 0.0004
- LoRA rank: 16
## Contribute your own examples
You can use the [community tab](https://huggingface.co/BootesVoid/cme5srgxo070i6aq1xn74g11g_cme948f8h03y5rts87cbphbej/discussions) to add images that show off what you’ve made with this LoRA.
|
mveroe/Qwen2.5-1.5B_1024_1_1p0_0p0_1p0_sft
|
mveroe
| 2025-08-12T22:57:34Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"conversational",
"base_model:Qwen/Qwen2.5-1.5B",
"base_model:finetune:Qwen/Qwen2.5-1.5B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T22:53:59Z
|
---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen2.5-1.5B
tags:
- generated_from_trainer
model-index:
- name: Qwen2.5-1.5B_1024_1_1p0_0p0_1p0_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. -->
# Qwen2.5-1.5B_1024_1_1p0_0p0_1p0_sft
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B](https://huggingface.co/Qwen/Qwen2.5-1.5B) 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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 32
- optimizer: Use OptimizerNames.ADAFACTOR and the args are:
No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
### Training results
### Framework versions
- Transformers 4.55.0
- Pytorch 2.8.0a0+5228986c39.nv25.06
- Datasets 4.0.0
- Tokenizers 0.21.4
|
ociolli/ape_LR
|
ociolli
| 2025-08-12T22:55:27Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"generated_from_trainer",
"conversational",
"base_model:yzhuang/Llama-3.1-8B-Instruct-attn",
"base_model:finetune:yzhuang/Llama-3.1-8B-Instruct-attn",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T10:00:18Z
|
---
library_name: transformers
base_model: yzhuang/Llama-3.1-8B-Instruct-attn
tags:
- generated_from_trainer
model-index:
- name: ape_LR
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. -->
# ape_LR
This model is a fine-tuned version of [yzhuang/Llama-3.1-8B-Instruct-attn](https://huggingface.co/yzhuang/Llama-3.1-8B-Instruct-attn) 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: 0.0005
- train_batch_size: 1
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- total_eval_batch_size: 4
- 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: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 3
### Training results
### Framework versions
- Transformers 4.55.0
- Pytorch 2.8.0+cu128
- Datasets 3.6.0
- Tokenizers 0.21.4
|
racro/qwen2-7b-instruct-trl-sft-ChartQA
|
racro
| 2025-08-12T22:51:41Z
| 0
| 0
|
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:Qwen/Qwen2-VL-7B-Instruct",
"base_model:finetune:Qwen/Qwen2-VL-7B-Instruct",
"endpoints_compatible",
"region:us"
] | null | 2025-08-11T20:58:36Z
|
---
base_model: Qwen/Qwen2-VL-7B-Instruct
library_name: transformers
model_name: qwen2-7b-instruct-trl-sft-ChartQA
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for qwen2-7b-instruct-trl-sft-ChartQA
This model is a fine-tuned version of [Qwen/Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-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="racro/qwen2-7b-instruct-trl-sft-ChartQA", 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/roongta-ritik20-new-york-university/roongta-ritik20-new-york-university/runs/2dxihegu)
This model was trained with SFT.
### Framework versions
- TRL: 0.22.0.dev0
- Transformers: 4.56.0.dev0
- Pytorch: 2.5.1+cu121
- Datasets: 4.0.0
- Tokenizers: 0.21.2
## 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}}
}
```
|
indoempatnol/blockassist-bc-fishy_wary_swan_1755037513
|
indoempatnol
| 2025-08-12T22:50:53Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"fishy wary swan",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:50:49Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- fishy wary swan
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
jimanex/blockassist-bc-rangy_peaceful_stingray_1755038822
|
jimanex
| 2025-08-12T22:48:50Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"rangy peaceful stingray",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:48:46Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- rangy peaceful stingray
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
Ammar-Zen/blockassist-bc-shrewd_wild_turkey_1755033561
|
Ammar-Zen
| 2025-08-12T22:48:27Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"shrewd wild turkey",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:48:18Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- shrewd wild turkey
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
SeacowX/olmoe-7x1b-melacious-code-E1
|
SeacowX
| 2025-08-12T22:46:15Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"olmoe",
"text-generation",
"arxiv:1910.09700",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T21:59:30Z
|
---
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]
|
jaayg23/phi3-mini-med-adapter
|
jaayg23
| 2025-08-12T22:43:40Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:microsoft/Phi-3-mini-4k-instruct",
"base_model:finetune:microsoft/Phi-3-mini-4k-instruct",
"endpoints_compatible",
"region:us"
] | null | 2025-08-12T22:32:25Z
|
---
base_model: microsoft/Phi-3-mini-4k-instruct
library_name: transformers
model_name: phi3-mini-med-adapter
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for phi3-mini-med-adapter
This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-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="jaayg23/phi3-mini-med-adapter", 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: 0.12.1
- Transformers: 4.46.2
- Pytorch: 2.6.0+cu124
- Datasets: 3.1.0
- Tokenizers: 0.20.3
## 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}}
}
```
|
aksamlan/blockassist-bc-miniature_roaring_python_1755038551
|
aksamlan
| 2025-08-12T22:43:11Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"miniature roaring python",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:43:06Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- miniature roaring python
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
MaestroDev19/gemma-3-cyber-expert-2.1
|
MaestroDev19
| 2025-08-12T22:41:12Z
| 0
| 0
|
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:google/gemma-3-1b-pt",
"base_model:finetune:google/gemma-3-1b-pt",
"endpoints_compatible",
"region:us"
] | null | 2025-08-12T21:43:27Z
|
---
base_model: google/gemma-3-1b-pt
library_name: transformers
model_name: gemma-3-cyber-expert-2.1
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for gemma-3-cyber-expert-2.1
This model is a fine-tuned version of [google/gemma-3-1b-pt](https://huggingface.co/google/gemma-3-1b-pt).
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="MaestroDev19/gemma-3-cyber-expert-2.1", 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: 0.15.2
- Transformers: 4.55.0
- Pytorch: 2.6.0+cu124
- Datasets: 3.3.2
- Tokenizers: 0.21.4
## 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}}
}
```
|
ypszn/blockassist-bc-yapping_pawing_worm_1755038198
|
ypszn
| 2025-08-12T22:37:35Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"yapping pawing worm",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:37:31Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- yapping pawing worm
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
OneAfterlife/upscale_models
|
OneAfterlife
| 2025-08-12T22:37:23Z
| 0
| 0
| null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-08-12T21:07:50Z
|
---
license: apache-2.0
---
|
BienKieu/codet5p_multi_dual
|
BienKieu
| 2025-08-12T22:36:48Z
| 0
| 0
|
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"arxiv:1910.09700",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | null | 2025-08-12T22:35:48Z
|
---
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]
|
BootesVoid/cme5srgxo070i6aq1xn74g11g_cme93516i03v0rts8jj5jamv8
|
BootesVoid
| 2025-08-12T22:35:38Z
| 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
| 2025-08-12T22:35:36Z
|
---
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: BLOND
---
# Cme5Srgxo070I6Aq1Xn74G11G_Cme93516I03V0Rts8Jj5Jamv8
<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 `BLOND` to trigger the image generation.
## Run this LoRA with an API using Replicate
```py
import replicate
input = {
"prompt": "BLOND",
"lora_weights": "https://huggingface.co/BootesVoid/cme5srgxo070i6aq1xn74g11g_cme93516i03v0rts8jj5jamv8/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('BootesVoid/cme5srgxo070i6aq1xn74g11g_cme93516i03v0rts8jj5jamv8', weight_name='lora.safetensors')
image = pipeline('BLOND').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: 2000
- Learning rate: 0.0004
- LoRA rank: 16
## Contribute your own examples
You can use the [community tab](https://huggingface.co/BootesVoid/cme5srgxo070i6aq1xn74g11g_cme93516i03v0rts8jj5jamv8/discussions) to add images that show off what you’ve made with this LoRA.
|
ggozzy/blockassist-bc-stubby_yapping_mandrill_1755038030
|
ggozzy
| 2025-08-12T22:35:34Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"stubby yapping mandrill",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:35:17Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- stubby yapping mandrill
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
gensynnode2025/blockassist-bc-clawed_humming_mole_1755037829
|
gensynnode2025
| 2025-08-12T22:34:44Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"clawed humming mole",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:31:41Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- clawed humming mole
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
k1000dai/residualact_libero_allaction
|
k1000dai
| 2025-08-12T22:34:27Z
| 0
| 0
|
lerobot
|
[
"lerobot",
"safetensors",
"robotics",
"residualact",
"dataset:k1000dai/libero_smolvla_allaction",
"license:apache-2.0",
"region:us"
] |
robotics
| 2025-08-12T22:34:15Z
|
---
datasets: k1000dai/libero_smolvla_allaction
library_name: lerobot
license: apache-2.0
model_name: residualact
pipeline_tag: robotics
tags:
- robotics
- residualact
- lerobot
---
# Model Card for residualact
<!-- Provide a quick summary of what the model is/does. -->
_Model type not recognized — please update this template._
This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).
See the full documentation at [LeRobot Docs](https://huggingface.co/docs/lerobot/index).
---
## How to Get Started with the Model
For a complete walkthrough, see the [training guide](https://huggingface.co/docs/lerobot/il_robots#train-a-policy).
Below is the short version on how to train and run inference/eval:
### Train from scratch
```bash
python -m lerobot.scripts.train \
--dataset.repo_id=${HF_USER}/<dataset> \
--policy.type=act \
--output_dir=outputs/train/<desired_policy_repo_id> \
--job_name=lerobot_training \
--policy.device=cuda \
--policy.repo_id=${HF_USER}/<desired_policy_repo_id>
--wandb.enable=true
```
_Writes checkpoints to `outputs/train/<desired_policy_repo_id>/checkpoints/`._
### Evaluate the policy/run inference
```bash
python -m lerobot.record \
--robot.type=so100_follower \
--dataset.repo_id=<hf_user>/eval_<dataset> \
--policy.path=<hf_user>/<desired_policy_repo_id> \
--episodes=10
```
Prefix the dataset repo with **eval\_** and supply `--policy.path` pointing to a local or hub checkpoint.
---
## Model Details
- **License:** apache-2.0
|
utkata31/blockassist-bc-singing_deft_okapi_1755037957
|
utkata31
| 2025-08-12T22:33:25Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"singing deft okapi",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:33:17Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- singing deft okapi
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
ps2program/gpt2-finetuned-ps2prahlad
|
ps2program
| 2025-08-12T22:30:14Z
| 0
| 0
| null |
[
"safetensors",
"gpt2",
"8-bit",
"bitsandbytes",
"region:us"
] | null | 2025-08-12T22:09:17Z
|
## Model Card for GPT-2 Fine-tuned with LoRA
### Model Details
#### Model Description
This model is a GPT-2 small model that has been fine-tuned using LoRA adapters. The fine-tuning was performed on a curated, small storytelling dataset, such as TinyStories, to enhance the model's ability to generate coherent and creative text in that domain.
- **Developed by:** Prahlad Sahu (ps2program)
- **Shared by:** Prahlad Sahu (ps2program)
- **Model type:** Causal Language Model
- **Language(s) (NLP):** English
- **License:** Apache 2.0
- **Finetuned from model:** GPT-2 small
#### Model Sources
- **Repository:** [https://huggingface.co/ps2program/gpt2-finetuned-ps2prahlad](https://huggingface.co/ps2program/gpt2-finetuned-ps2prahlad)
- **Paper [optional]:** Radford et al., 2019
### Uses
#### Direct Use
This model is intended for **text generation tasks**, specifically for creating stories, creative writing, and other forms of narrative content. It can be used directly to generate text from a given prompt.
#### Downstream Use [optional]
The model can serve as a **starting point for further fine-tuning** on domain-specific tasks. For example, it could be adapted for generating scripts, poetry, or other types of creative text by training it on a relevant dataset.
#### Out-of-Scope Use
This model is **not suitable for factual or safety-critical applications** because it may produce biased, nonsensical, or inaccurate content. It should not be used for tasks requiring factual accuracy, such as generating news articles, medical advice, or legal documents.
### Bias, Risks, and Limitations
This model, like many large language models, may produce **biased, nonsensical, or inappropriate content**. Its outputs are heavily dependent on the quality and size of the fine-tuning dataset, which could introduce biases present in the data.
#### Recommendations
Users should be aware of the model's limitations and verify any content it produces for accuracy and appropriateness. It is recommended to use the model in a controlled environment and for its intended creative purposes only.
### How to Get Started with the Model
To get started, you can use the Hugging Face `transformers` library. The following Python code demonstrates how to load and use the model for text generation:
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
model_name = "ps2program/gpt2-finetuned-ps2prahlad"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
prompt = "Once upon a time"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=50, do_sample=True, temperature=0.7)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
### Training Details
#### Training Data
The model was fine-tuned on the **TinyStories** dataset or a similar small text dataset curated for storytelling.
#### Training Procedure
The fine-tuning was performed using **LoRA (Low-Rank Adaptation)** adapters, which are a parameter-efficient fine-tuning method. The adapters were then merged into the base GPT-2 model.
- **Training regime:** Not specified, but typically involves mixed precision training (e.g., `fp16`).
### Evaluation
#### Testing Data, Factors & Metrics
- **Testing Data:** The model was likely evaluated on a held-out portion of the TinyStories dataset.
- **Metrics:** The primary metric used for evaluation was **perplexity**, which measures how well the model predicts a sample of text. The reported perplexity value is 12.34.
### Model Card Contact
For questions or feedback regarding this model card, please contact ps2program (Prahlad Sahu).
|
mradermacher/Jan-v1-4B-i1-GGUF
|
mradermacher
| 2025-08-12T22:28:18Z
| 0
| 0
|
transformers
|
[
"transformers",
"gguf",
"en",
"base_model:janhq/Jan-v1-4B",
"base_model:quantized:janhq/Jan-v1-4B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2025-08-12T19:09:22Z
|
---
base_model: janhq/Jan-v1-4B
language:
- en
library_name: transformers
license: apache-2.0
mradermacher:
readme_rev: 1
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
<!-- ### quants: Q2_K IQ3_M Q4_K_S IQ3_XXS Q3_K_M small-IQ4_NL Q4_K_M IQ2_M Q6_K IQ4_XS Q2_K_S IQ1_M Q3_K_S IQ2_XXS Q3_K_L IQ2_XS Q5_K_S IQ2_S IQ1_S Q5_K_M Q4_0 IQ3_XS Q4_1 IQ3_S -->
<!-- ### quants_skip: -->
<!-- ### skip_mmproj: -->
weighted/imatrix quants of https://huggingface.co/janhq/Jan-v1-4B
<!-- provided-files -->
***For a convenient overview and download list, visit our [model page for this model](https://hf.tst.eu/model#Jan-v1-4B-i1-GGUF).***
static quants are available at https://huggingface.co/mradermacher/Jan-v1-4B-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/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ1_S.gguf) | i1-IQ1_S | 1.2 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ1_M.gguf) | i1-IQ1_M | 1.2 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.3 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 1.5 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ2_S.gguf) | i1-IQ2_S | 1.5 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ2_M.gguf) | i1-IQ2_M | 1.6 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 1.7 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q2_K.gguf) | i1-Q2_K | 1.8 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 1.8 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 1.9 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 2.0 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ3_S.gguf) | i1-IQ3_S | 2.0 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ3_M.gguf) | i1-IQ3_M | 2.1 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 2.2 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 2.3 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 2.4 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q4_0.gguf) | i1-Q4_0 | 2.5 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-IQ4_NL.gguf) | i1-IQ4_NL | 2.5 | prefer IQ4_XS |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 2.5 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 2.6 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q4_1.gguf) | i1-Q4_1 | 2.7 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 2.9 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 3.0 | |
| [GGUF](https://huggingface.co/mradermacher/Jan-v1-4B-i1-GGUF/resolve/main/Jan-v1-4B.i1-Q6_K.gguf) | i1-Q6_K | 3.4 | 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 -->
|
ypszn/blockassist-bc-yapping_pawing_worm_1755037431
|
ypszn
| 2025-08-12T22:25:06Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"yapping pawing worm",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:25:01Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- yapping pawing worm
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
mang3dd/blockassist-bc-tangled_slithering_alligator_1755035878
|
mang3dd
| 2025-08-12T22:24:21Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"tangled slithering alligator",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:24:17Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- tangled slithering alligator
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
Dejiat/blockassist-bc-savage_unseen_bobcat_1755037314
|
Dejiat
| 2025-08-12T22:22:28Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"savage unseen bobcat",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:22:20Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- savage unseen bobcat
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
indoempatnol/blockassist-bc-fishy_wary_swan_1755035808
|
indoempatnol
| 2025-08-12T22:21:59Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"fishy wary swan",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:21:56Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- fishy wary swan
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
DevQuasar/inclusionAI.ASearcher-Local-7B-GGUF
|
DevQuasar
| 2025-08-12T22:19:17Z
| 0
| 0
| null |
[
"gguf",
"text-generation",
"base_model:inclusionAI/ASearcher-Local-7B",
"base_model:quantized:inclusionAI/ASearcher-Local-7B",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-08-12T21:35:55Z
|
---
base_model:
- inclusionAI/ASearcher-Local-7B
pipeline_tag: text-generation
---
[<img src="https://raw.githubusercontent.com/csabakecskemeti/devquasar/main/dq_logo_black-transparent.png" width="200"/>](https://devquasar.com)
Quantized version of: [inclusionAI/ASearcher-Local-7B](https://huggingface.co/inclusionAI/ASearcher-Local-7B)
'Make knowledge free for everyone'
<p align="center">
Made with <br>
<a href="https://www.civo.com/" target="_blank">
<img src="https://www.civo.com/assets/public/brand-assets/civo-logo-colour-60cc1622dedf346f7afde1fff760523f731b0aac106a5465af98ff4073114b74.svg" width="100"/>
</a>
</p>
<a href='https://ko-fi.com/L4L416YX7C' target='_blank'><img height='36' style='border:0px;height:36px;' src='https://storage.ko-fi.com/cdn/kofi6.png?v=6' border='0' alt='Buy Me a Coffee at ko-fi.com' /></a>
|
tensorblock/tarun7r_Finance-Llama-8B-GGUF
|
tensorblock
| 2025-08-12T22:17:31Z
| 0
| 0
|
transformers
|
[
"transformers",
"gguf",
"text-generation-inference",
"finance",
"economics",
"TensorBlock",
"GGUF",
"text-generation",
"en",
"dataset:Josephgflowers/Finance-Instruct-500k",
"base_model:tarun7r/Finance-Llama-8B",
"base_model:quantized:tarun7r/Finance-Llama-8B",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-08-12T22:17:21Z
|
---
license: apache-2.0
tags:
- text-generation-inference
- finance
- economics
- TensorBlock
- GGUF
datasets:
- Josephgflowers/Finance-Instruct-500k
language:
- en
base_model: tarun7r/Finance-Llama-8B
pipeline_tag: text-generation
library_name: transformers
---
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
[](https://tensorblock.co)
[](https://twitter.com/tensorblock_aoi)
[](https://discord.gg/Ej5NmeHFf2)
[](https://github.com/TensorBlock)
[](https://t.me/TensorBlock)
## tarun7r/Finance-Llama-8B - GGUF
<div style="text-align: left; margin: 20px 0;">
<a href="https://discord.com/invite/Ej5NmeHFf2" style="display: inline-block; padding: 10px 20px; background-color: #5865F2; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
Join our Discord to learn more about what we're building ↗
</a>
</div>
This repo contains GGUF format model files for [tarun7r/Finance-Llama-8B](https://huggingface.co/tarun7r/Finance-Llama-8B).
The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b5753](https://github.com/ggml-org/llama.cpp/commit/73e53dc834c0a2336cd104473af6897197b96277).
## Our projects
<table border="1" cellspacing="0" cellpadding="10">
<tr>
<th colspan="2" style="font-size: 25px;">Forge</th>
</tr>
<tr>
<th colspan="2">
<img src="https://imgur.com/faI5UKh.jpeg" alt="Forge Project" width="900"/>
</th>
</tr>
<tr>
<th colspan="2">An OpenAI-compatible multi-provider routing layer.</th>
</tr>
<tr>
<th colspan="2">
<a href="https://github.com/TensorBlock/forge" target="_blank" style="
display: inline-block;
padding: 8px 16px;
background-color: #FF7F50;
color: white;
text-decoration: none;
border-radius: 6px;
font-weight: bold;
font-family: sans-serif;
">🚀 Try it now! 🚀</a>
</th>
</tr>
<tr>
<th style="font-size: 25px;">Awesome MCP Servers</th>
<th style="font-size: 25px;">TensorBlock Studio</th>
</tr>
<tr>
<th><img src="https://imgur.com/2Xov7B7.jpeg" alt="MCP Servers" width="450"/></th>
<th><img src="https://imgur.com/pJcmF5u.jpeg" alt="Studio" width="450"/></th>
</tr>
<tr>
<th>A comprehensive collection of Model Context Protocol (MCP) servers.</th>
<th>A lightweight, open, and extensible multi-LLM interaction studio.</th>
</tr>
<tr>
<th>
<a href="https://github.com/TensorBlock/awesome-mcp-servers" target="_blank" style="
display: inline-block;
padding: 8px 16px;
background-color: #FF7F50;
color: white;
text-decoration: none;
border-radius: 6px;
font-weight: bold;
font-family: sans-serif;
">👀 See what we built 👀</a>
</th>
<th>
<a href="https://github.com/TensorBlock/TensorBlock-Studio" target="_blank" style="
display: inline-block;
padding: 8px 16px;
background-color: #FF7F50;
color: white;
text-decoration: none;
border-radius: 6px;
font-weight: bold;
font-family: sans-serif;
">👀 See what we built 👀</a>
</th>
</tr>
</table>
## Prompt template
```
Unable to determine prompt format automatically. Please check the original model repository for the correct prompt format.
```
## Model file specification
| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [Finance-Llama-8B-Q2_K.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q2_K.gguf) | Q2_K | 0.008 GB | smallest, significant quality loss - not recommended for most purposes |
| [Finance-Llama-8B-Q3_K_S.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q3_K_S.gguf) | Q3_K_S | 0.008 GB | very small, high quality loss |
| [Finance-Llama-8B-Q3_K_M.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q3_K_M.gguf) | Q3_K_M | 0.008 GB | very small, high quality loss |
| [Finance-Llama-8B-Q3_K_L.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q3_K_L.gguf) | Q3_K_L | 0.008 GB | small, substantial quality loss |
| [Finance-Llama-8B-Q4_0.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q4_0.gguf) | Q4_0 | 0.008 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [Finance-Llama-8B-Q4_K_S.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q4_K_S.gguf) | Q4_K_S | 0.008 GB | small, greater quality loss |
| [Finance-Llama-8B-Q4_K_M.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q4_K_M.gguf) | Q4_K_M | 0.008 GB | medium, balanced quality - recommended |
| [Finance-Llama-8B-Q5_0.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q5_0.gguf) | Q5_0 | 0.008 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [Finance-Llama-8B-Q5_K_S.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q5_K_S.gguf) | Q5_K_S | 0.008 GB | large, low quality loss - recommended |
| [Finance-Llama-8B-Q5_K_M.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q5_K_M.gguf) | Q5_K_M | 0.008 GB | large, very low quality loss - recommended |
| [Finance-Llama-8B-Q6_K.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q6_K.gguf) | Q6_K | 0.008 GB | very large, extremely low quality loss |
| [Finance-Llama-8B-Q8_0.gguf](https://huggingface.co/tensorblock/tarun7r_Finance-Llama-8B-GGUF/blob/main/Finance-Llama-8B-Q8_0.gguf) | Q8_0 | 0.008 GB | very large, extremely low quality loss - not recommended |
## Downloading instruction
### Command line
Firstly, install Huggingface Client
```shell
pip install -U "huggingface_hub[cli]"
```
Then, downoad the individual model file the a local directory
```shell
huggingface-cli download tensorblock/tarun7r_Finance-Llama-8B-GGUF --include "Finance-Llama-8B-Q2_K.gguf" --local-dir MY_LOCAL_DIR
```
If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:
```shell
huggingface-cli download tensorblock/tarun7r_Finance-Llama-8B-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
|
EYEDOL/FROM_C3_2
|
EYEDOL
| 2025-08-12T22:14:16Z
| 0
| 0
|
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"whisper",
"automatic-speech-recognition",
"hf-asr-leaderboard",
"generated_from_trainer",
"sw",
"dataset:mozilla-foundation/common_voice_11_0",
"base_model:EYEDOL/FROM_C3_1",
"base_model:finetune:EYEDOL/FROM_C3_1",
"model-index",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2025-08-12T22:13:59Z
|
---
library_name: transformers
language:
- sw
base_model: EYEDOL/FROM_C3_1
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
model-index:
- name: ASR_FROM_C3
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 11.0
type: mozilla-foundation/common_voice_11_0
config: sw
split: None
args: 'config: sw, split: test'
metrics:
- name: Wer
type: wer
value: 17.02991010874313
---
<!-- 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. -->
# ASR_FROM_C3
This model is a fine-tuned version of [EYEDOL/FROM_C3_1](https://huggingface.co/EYEDOL/FROM_C3_1) on the Common Voice 11.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2333
- Wer: 17.0299
## 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-05
- train_batch_size: 16
- 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
- lr_scheduler_warmup_steps: 500
- num_epochs: 1
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:-------:|
| 0.0582 | 0.8684 | 2000 | 0.2333 | 17.0299 |
### Framework versions
- Transformers 4.52.4
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.2
|
MaestroDev19/gemma-3-cyber-expert
|
MaestroDev19
| 2025-08-12T22:13:22Z
| 0
| 0
|
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:google/gemma-3-1b-pt",
"base_model:finetune:google/gemma-3-1b-pt",
"endpoints_compatible",
"region:us"
] | null | 2025-08-12T20:27:02Z
|
---
base_model: google/gemma-3-1b-pt
library_name: transformers
model_name: gemma-3-cyber-expert
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for gemma-3-cyber-expert
This model is a fine-tuned version of [google/gemma-3-1b-pt](https://huggingface.co/google/gemma-3-1b-pt).
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="MaestroDev19/gemma-3-cyber-expert", 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: 0.15.2
- Transformers: 4.55.0
- Pytorch: 2.6.0+cu124
- Datasets: 3.3.2
- Tokenizers: 0.21.4
## 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}}
}
```
|
kojeklollipop/blockassist-bc-spotted_amphibious_stork_1755035235
|
kojeklollipop
| 2025-08-12T22:12:34Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"spotted amphibious stork",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:12:30Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- spotted amphibious stork
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
ypszn/blockassist-bc-yapping_pawing_worm_1755036698
|
ypszn
| 2025-08-12T22:12:33Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"yapping pawing worm",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:12:29Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- yapping pawing worm
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
jmartin233/dqn-SpaceInvadersNoFrameskip-v4
|
jmartin233
| 2025-08-12T22:11:40Z
| 0
| 0
|
stable-baselines3
|
[
"stable-baselines3",
"SpaceInvadersNoFrameskip-v4",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2025-08-12T22:11:01Z
|
---
library_name: stable-baselines3
tags:
- SpaceInvadersNoFrameskip-v4
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: DQN
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: SpaceInvadersNoFrameskip-v4
type: SpaceInvadersNoFrameskip-v4
metrics:
- type: mean_reward
value: 560.00 +/- 140.87
name: mean_reward
verified: false
---
# **DQN** Agent playing **SpaceInvadersNoFrameskip-v4**
This is a trained model of a **DQN** agent playing **SpaceInvadersNoFrameskip-v4**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3)
and the [RL Zoo](https://github.com/DLR-RM/rl-baselines3-zoo).
The RL Zoo is a training framework for Stable Baselines3
reinforcement learning agents,
with hyperparameter optimization and pre-trained agents included.
## Usage (with SB3 RL Zoo)
RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo<br/>
SB3: https://github.com/DLR-RM/stable-baselines3<br/>
SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib
SBX (SB3 + Jax): https://github.com/araffin/sbx
Install the RL Zoo (with SB3 and SB3-Contrib):
```bash
pip install rl_zoo3
```
```
# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga jmartin233 -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
If you installed the RL Zoo3 via pip (`pip install rl_zoo3`), from anywhere you can do:
```
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga jmartin233 -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
## Training (with the RL Zoo)
```
python -m rl_zoo3.train --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga jmartin233
```
## Hyperparameters
```python
OrderedDict([('batch_size', 32),
('buffer_size', 100000),
('env_wrapper',
['stable_baselines3.common.atari_wrappers.AtariWrapper']),
('exploration_final_eps', 0.01),
('exploration_fraction', 0.1),
('frame_stack', 4),
('gradient_steps', 1),
('learning_rate', 0.0001),
('learning_starts', 100000),
('n_timesteps', 1000000.0),
('optimize_memory_usage', False),
('policy', 'CnnPolicy'),
('target_update_interval', 1000),
('train_freq', 4),
('normalize', False)])
```
# Environment Arguments
```python
{'render_mode': 'rgb_array'}
```
|
gensynnode2025/blockassist-bc-clawed_humming_mole_1755036436
|
gensynnode2025
| 2025-08-12T22:11:27Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"clawed humming mole",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:08:00Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- clawed humming mole
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
jerrrycans/watermark20000x2
|
jerrrycans
| 2025-08-12T22:11:12Z
| 0
| 0
|
diffusers
|
[
"diffusers",
"flux",
"image-to-image",
"lora",
"replicate",
"base_model:black-forest-labs/FLUX.1-Kontext-dev",
"base_model:adapter:black-forest-labs/FLUX.1-Kontext-dev",
"license:other",
"region:us"
] |
image-to-image
| 2025-08-12T21:13:05Z
|
---
license: other
license_name: flux1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev/blob/main/LICENSE.md
tags:
- flux
- image-to-image
- lora
- diffusers
- replicate
base_model: black-forest-labs/FLUX.1-Kontext-dev
pipeline_tag: image-to-image
# widget:
# - src: https://...
# text: >-
# prompt
# output:
# url: https://...
instance_prompt: remove all the watermarks from this image, all watermarks that are over this image, text watermarks, logo watermarks etc
---
# Watermark20000X2
<Gallery />
## About this LoRA
This is a [LoRA](https://replicate.com/docs/guides/working-with-loras) for the FLUX.1-Kontext-dev image-to-image model. It can be used with diffusers or ComfyUI.
It was trained on [Replicate](https://replicate.com/) using: https://replicate.com/replicate/fast-flux-kontext-trainer/train
## Prompt instruction
You should use `remove all the watermarks from this image, all watermarks that are over this image, text watermarks, logo watermarks etc` as part of the prompt instruction for your image-to-image editing.
## Training details
- Steps: 20000
- Learning rate: 0.001
- LoRA rank: 16
## Contribute your own examples
You can use the [community tab](https://huggingface.co/jerrrycans/watermark20000x2/discussions) to add images that show off what you’ve made with this LoRA.
|
ggozzy/blockassist-bc-stubby_yapping_mandrill_1755036401
|
ggozzy
| 2025-08-12T22:08:08Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"stubby yapping mandrill",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:07:53Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- stubby yapping mandrill
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
nicoboss/dbrx-base
|
nicoboss
| 2025-08-12T22:07:54Z
| 0
| 0
| null |
[
"safetensors",
"dbrx",
"custom_code",
"arxiv:2211.15841",
"arxiv:2304.11277",
"license:other",
"region:us"
] | null | 2025-08-12T22:02:39Z
|
---
extra_gated_heading: You need to share contact information with Databricks to access this model
extra_gated_prompt: >-
### DBRX Terms of Use
Use of DBRX is governed by the [Databricks Open Model License](https://www.databricks.com/legal/open-model-license) and the [Databricks Open Model Acceptable Use Policy](https://www.databricks.com/legal/acceptable-use-policy-open-model).
extra_gated_fields:
First Name: text
Last Name: text
Organization: text
Purpose for Base Model Access: text
By clicking 'Submit' below, I accept the terms of the license and acknowledge that the information I provide will be collected, stored, processed, and shared in accordance with Databricks' Privacy Notice and I understand I can update my preferences at any time: checkbox
extra_gated_description: >-
The information you provide will be collected, stored, processed, and shared in accordance with Databricks [Privacy Notice](https://www.databricks.com/legal/privacynotice).
extra_gated_button_content: Submit
inference: false
license: other
license_name: databricks-open-model-license
license_link: https://www.databricks.com/legal/open-model-license
---
# llama.cpp fixed tokenizer
llama.cpp fixed tokenizer applied from https://huggingface.co/treehugg3/dbrx-base-tokenizer-llamacpp which in turn is based on https://huggingface.co/LnL-AI/dbrx-base-tokenizer
# Re-upload because original repo is gated
Don't do that shit. Come on. Open weights mean open weights. Not gate.
# DBRX Base
* DBRX Base is a mixture-of-experts (MoE) large language model trained from scratch by Databricks.
* We are releasing both DBRX Base, a pretrained base model, and DBRX Instruct, a fine-tuned version for few-turn interactions, under [an open license](https://www.databricks.com/legal/open-model-license).
* This is the repository for DBRX Base. DBRX Instruct can be found [here](https://huggingface.co/databricks/dbrx-instruct).
* For full details on the DBRX models, please read our [technical blog post](https://www.databricks.com/blog/introducing-dbrx-new-state-art-open-llm).
## Model Overview
DBRX is a [transformer-based](https://www.isattentionallyouneed.com/) decoder-only large language model (LLM) that was trained using next-token prediction.
It uses a *fine-grained* mixture-of-experts (MoE) architecture with 132B total parameters of which 36B parameters are active on any input.
It was pre-trained on 12T tokens of text and code data.
Compared to other open MoE models like Mixtral-8x7B and Grok-1, DBRX is fine-grained, meaning it uses a larger number of smaller experts. DBRX has 16 experts and chooses 4, while Mixtral-8x7B and Grok-1 have 8 experts and choose 2.
This provides 65x more possible combinations of experts and we found that this improves model quality.
DBRX uses rotary position encodings (RoPE), gated linear units (GLU), and grouped query attention (GQA).
It uses the GPT-4 tokenizer as provided in the [tiktoken](https://github.com/openai/tiktoken) repository.
We made these choices based on exhaustive evaluation and scaling experiments.
DBRX was pretrained on 12T tokens of carefully curated data and a maximum context length of 32K tokens.
We estimate that this data is at least 2x better token-for-token than the data we used to pretrain the MPT family of models.
This new dataset was developed using the full suite of Databricks tools, including Apache Spark™ and Databricks notebooks for data processing, and Unity Catalog for data management and governance.
We used curriculum learning for pretraining, changing the data mix during training in ways we found to substantially improve model quality.
* **Inputs:** DBRX only accepts text-based inputs and accepts a context length of up to 32768 tokens.
* **Outputs:** DBRX only produces text-based outputs.
* **Model Architecture:** More detailed information about DBRX Instruct and DBRX Base can be found in our [technical blog post](https://www.databricks.com/blog/introducing-dbrx-new-state-art-open-llm).
* **License:** [Databricks Open Model License](https://www.databricks.com/legal/open-model-license)
* **Acceptable Use Policy:** [Databricks Open Model Acceptable Use Policy](https://www.databricks.com/legal/acceptable-use-policy-open-model)
* **Version:** 1.0
* **Owner:** Databricks, Inc.
## Usage
These are several general ways to use the DBRX models:
* DBRX Base and DBRX Instruct are available for download on HuggingFace (see our Quickstart guide below). This is the HF repository for DBRX Base; DBRX Instruct can be found [here](https://huggingface.co/databricks/dbrx-instruct).
* The DBRX model repository can be found on GitHub [here](https://github.com/databricks/dbrx).
* DBRX Base and DBRX Instruct are available with [Databricks Foundation Model APIs](https://docs.databricks.com/en/machine-learning/foundation-models/index.html) via both *Pay-per-token* and *Provisioned Throughput* endpoints. These are enterprise-ready deployments.
* For more information on how to fine-tune using LLM-Foundry, please take a look at our LLM pretraining and fine-tuning [documentation](https://github.com/mosaicml/llm-foundry/blob/main/scripts/train/README.md).
## Quickstart Guide
**NOTE: This is DBRX Base, and has not been instruction finetuned. It has not been trained for interactive chat and is only a completion model.**
If you are looking for the finetuned model, please use [DBRX Instruct](https://huggingface.co/databricks/dbrx-instruct).
Getting started with DBRX models is easy with the `transformers` library. The model requires ~264GB of RAM and the following packages:
```bash
pip install "transformers>=4.39.2" "tiktoken>=0.6.0"
```
If you'd like to speed up download time, you can use the `hf_transfer` package as described by Huggingface [here](https://huggingface.co/docs/huggingface_hub/en/guides/download#faster-downloads).
```bash
pip install hf_transfer
export HF_HUB_ENABLE_HF_TRANSFER=1
```
You will need to request access to this repository to download the model. Once this is granted,
[obtain an access token](https://huggingface.co/docs/hub/en/security-tokens) with `read` permission, and supply the token below.
### Run the model on a CPU:
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
tokenizer = AutoTokenizer.from_pretrained("Undi95/dbrx-base", trust_remote_code=True, token="hf_YOUR_TOKEN")
model = AutoModelForCausalLM.from_pretrained("Undi95/dbrx-base", device_map="cpu", torch_dtype=torch.bfloat16, trust_remote_code=True, token="hf_YOUR_TOKEN")
input_text = "Databricks was founded in "
input_ids = tokenizer(input_text, return_tensors="pt")
outputs = model.generate(**input_ids, max_new_tokens=100)
print(tokenizer.decode(outputs[0]))
```
### Run the model on multiple GPUs:
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
tokenizer = AutoTokenizer.from_pretrained("Undi95/dbrx-base", trust_remote_code=True, token="hf_YOUR_TOKEN")
model = AutoModelForCausalLM.from_pretrained("Undi95/dbrx-base", device_map="auto", torch_dtype=torch.bfloat16, trust_remote_code=True, token="hf_YOUR_TOKEN")
input_text = "Databricks was founded in "
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")
outputs = model.generate(**input_ids, max_new_tokens=100)
print(tokenizer.decode(outputs[0]))
```
If your GPU system supports [FlashAttention2](https://huggingface.co/docs/transformers/perf_infer_gpu_one#flashattention-2), you can add `attn_implementation=”flash_attention_2”` as a keyword to `AutoModelForCausalLM.from_pretrained()` to achieve faster inference.
## Limitations and Ethical Considerations
### Training Dataset Limitations
The DBRX models were trained on 12T tokens of text, with a knowledge cutoff date of December 2023.
The training mix used for DBRX contains both natural-language and code examples. The vast majority of our training data is in the English language. We did not test DBRX for non-English proficiency. Therefore, DBRX should be considered a generalist model for text-based use in the English language.
DBRX does not have multimodal capabilities.
### Associated Risks and Recommendations
All foundation models are novel technologies that carry various risks, and may output information that is inaccurate, incomplete, biased, or offensive.
Users should exercise judgment and evaluate such output for accuracy and appropriateness for their desired use case before using or sharing it.
Databricks recommends [using retrieval augmented generation (RAG)](https://www.databricks.com/glossary/retrieval-augmented-generation-rag) in scenarios where accuracy and fidelity are important.
We also recommend that anyone using or fine-tuning either DBRX Base or DBRX Instruct perform additional testing around safety in the context of their particular application and domain.
## Intended Uses
### Intended Use Cases
The DBRX models are open, general-purpose LLMs intended and licensed for both commercial and research applications.
They can be further fine-tuned for various domain-specific natural language and coding tasks.
DBRX Base can be used as an off-the-shelf model for text completion for general English-language and coding tasks.
Please review the Associated Risks section above, as well as the [Databricks Open Model License](https://www.databricks.com/legal/open-model-license) and [Databricks Open Model Acceptable Use Policy](https://www.databricks.com/legal/acceptable-use-policy-open-model) for further information about permissible uses of DBRX Base and its derivatives.
### Out-of-Scope Use Cases
DBRX models are not intended to be used out-of-the-box in non-English languages and do not support native code execution, or other forms of function-calling.
DBRX models should not be used in any manner that violates applicable laws or regulations or in any other way that is prohibited by the [Databricks Open Model License](https://www.databricks.com/legal/open-model-license) and [Databricks Open Model Acceptable Use Policy](https://www.databricks.com/legal/acceptable-use-policy-open-model).
## Training Stack
MoE models are complicated to train, and the training of DBRX Base and DBRX Instruct was heavily supported by Databricks’ infrastructure for data processing and large-scale LLM training (e.g., [Composer](https://github.com/mosaicml/composer), [Streaming](https://github.com/mosaicml/streaming), [Megablocks](https://github.com/stanford-futuredata/megablocks), and [LLM Foundry](https://github.com/mosaicml/llm-foundry)).
Composer is our core library for large-scale training.
It provides an optimized training loop, easy [checkpointing](https://docs.mosaicml.com/projects/composer/en/latest/trainer/checkpointing.html) and [logging](https://docs.mosaicml.com/projects/composer/en/latest/trainer/logging.html#wood-logging),
[FSDP](https://pytorch.org/docs/stable/fsdp.html)-based [model sharding](https://docs.mosaicml.com/projects/composer/en/latest/notes/distributed_training.html#fullyshardeddataparallel-fsdp),
convenient [abstractions](https://docs.mosaicml.com/projects/composer/en/latest/trainer/time.html), extreme customizability via [callbacks](https://docs.mosaicml.com/projects/composer/en/latest/trainer/callbacks.html), and more.
Streaming enables fast, low cost, and scalable training on large datasets from cloud storage. It handles a variety of challenges around deterministic resumption as node counts change, avoiding redundant downloads across devices, high-quality shuffling at scale, sample-level random access, and speed.
Megablocks is a lightweight library for MoE training. Crucially, it supports “dropless MoE,” which avoids inefficient padding and is intended to provide deterministic outputs for a given sequence no matter what other sequences are in the batch.
LLM Foundry ties all of these libraries together to create a simple LLM pretraining, fine-tuning, and inference experience.
DBRX was trained using proprietary optimized versions of the above open source libraries, along with our [LLM training platform](https://www.databricks.com/product/machine-learning/mosaic-ai-training).
## Evaluation
We find that DBRX outperforms established open-source and open-weight base models on the [Databricks Model Gauntlet](https://www.databricks.com/blog/llm-evaluation-for-icl), the [Hugging Face Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard), and HumanEval.
The Databricks Model Gauntlet measures performance on more than 30 tasks across six categories: world knowledge, common sense reasoning, language understanding, reading comprehension, symbolic problem solving, and programming.
The Hugging Face Open LLM Leaderboard measures the average of ARC-Challenge, HellaSwag, MMLU, TruthfulQA, Winogrande and GSM8k.
HumanEval measures coding ability.
Full evaluation details can be found in our [technical blog post](https://www.databricks.com/blog/introducing-dbrx-new-state-art-open-llm).
## Acknowledgements
The DBRX models were made possible thanks in large part to the open-source community, especially:
* The [MegaBlocks](https://arxiv.org/abs/2211.15841) library, which established a foundation for our MoE implementation.
* [PyTorch FSDP](https://arxiv.org/abs/2304.11277), which we built on for distributed training.
|
chiachenwo/gpt-oss-20b-multilingual-reasoner
|
chiachenwo
| 2025-08-12T22:05:03Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"dataset:HuggingFaceH4/Multilingual-Thinking",
"base_model:openai/gpt-oss-20b",
"base_model:finetune:openai/gpt-oss-20b",
"endpoints_compatible",
"region:us"
] | null | 2025-08-09T21:34:15Z
|
---
base_model: openai/gpt-oss-20b
datasets: HuggingFaceH4/Multilingual-Thinking
library_name: transformers
model_name: gpt-oss-20b-multilingual-reasoner
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for gpt-oss-20b-multilingual-reasoner
This model is a fine-tuned version of [openai/gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) on the [HuggingFaceH4/Multilingual-Thinking](https://huggingface.co/datasets/HuggingFaceH4/Multilingual-Thinking) 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="chiachenwo/gpt-oss-20b-multilingual-reasoner", 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/chiachenwo-northwestern-university/huggingface/runs/2qptkklb)
This model was trained with SFT.
### Framework versions
- TRL: 0.21.0
- Transformers: 4.56.0.dev0
- Pytorch: 2.8.0
- Datasets: 4.0.0
- Tokenizers: 0.21.4
## 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}}
}
```
|
Sayemahsjn/blockassist-bc-playful_feline_octopus_1755034988
|
Sayemahsjn
| 2025-08-12T22:01:27Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"playful feline octopus",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T22:01:21Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- playful feline octopus
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
Aravindaxs/qwen_finetune
|
Aravindaxs
| 2025-08-12T22:00:22Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"qwen2_5_vl",
"image-to-text",
"text-generation-inference",
"unsloth",
"en",
"base_model:unsloth/Qwen2.5-VL-3B-Instruct",
"base_model:finetune:unsloth/Qwen2.5-VL-3B-Instruct",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
image-to-text
| 2025-08-12T21:57:29Z
|
---
base_model: unsloth/Qwen2.5-VL-3B-Instruct
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2_5_vl
license: apache-2.0
language:
- en
---
# Uploaded finetuned model
- **Developed by:** Aravindaxs
- **License:** apache-2.0
- **Finetuned from model :** unsloth/Qwen2.5-VL-3B-Instruct
This qwen2_5_vl 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)
|
Gemvision13/blockassist-bc-finicky_jagged_panda_1755035800
|
Gemvision13
| 2025-08-12T21:57:57Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"finicky jagged panda",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T21:57:49Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- finicky jagged panda
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
ggozzy/blockassist-bc-stubby_yapping_mandrill_1755035791
|
ggozzy
| 2025-08-12T21:57:57Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"stubby yapping mandrill",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T21:57:41Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- stubby yapping mandrill
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
Dejiat/blockassist-bc-savage_unseen_bobcat_1755035683
|
Dejiat
| 2025-08-12T21:55:56Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"savage unseen bobcat",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T21:55:50Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- savage unseen bobcat
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
anjalidutt/Deep-RL
|
anjalidutt
| 2025-08-12T21:55:55Z
| 0
| 0
|
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v3",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2025-08-12T21:55:35Z
|
---
library_name: stable-baselines3
tags:
- LunarLander-v3
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v3
type: LunarLander-v3
metrics:
- type: mean_reward
value: 251.56 +/- 11.48
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v3**
This is a trained model of a **PPO** agent playing **LunarLander-v3**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
mang3dd/blockassist-bc-tangled_slithering_alligator_1755034003
|
mang3dd
| 2025-08-12T21:54:39Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"tangled slithering alligator",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T21:54:33Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- tangled slithering alligator
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
indoempatnol/blockassist-bc-fishy_wary_swan_1755034061
|
indoempatnol
| 2025-08-12T21:53:35Z
| 0
| 0
| null |
[
"gensyn",
"blockassist",
"gensyn-blockassist",
"minecraft",
"fishy wary swan",
"arxiv:2504.07091",
"region:us"
] | null | 2025-08-12T21:53:32Z
|
---
tags:
- gensyn
- blockassist
- gensyn-blockassist
- minecraft
- fishy wary swan
---
# Gensyn BlockAssist
Gensyn's BlockAssist is a distributed extension of the paper [AssistanceZero: Scalably Solving Assistance Games](https://arxiv.org/abs/2504.07091).
|
arindam77op/medical_llama_model
|
arindam77op
| 2025-08-12T21:52:51Z
| 0
| 0
|
transformers
|
[
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-08-12T21:50:07Z
|
---
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]
|
Subsets and Splits
Filtered Qwen2.5 Distill Models
Identifies specific configurations of models by filtering cards that contain 'distill', 'qwen2.5', '7b' while excluding certain base models and incorrect model ID patterns, uncovering unique model variants.
Filtered Model Cards Count
Finds the count of entries with specific card details that include 'distill', 'qwen2.5', '7b' but exclude certain base models, revealing valuable insights about the dataset's content distribution.
Filtered Distill Qwen 7B Models
Filters for specific card entries containing 'distill', 'qwen', and '7b', excluding certain strings and patterns, to identify relevant model configurations.
Filtered Qwen-7b Model Cards
The query performs a detailed filtering based on specific keywords and excludes certain entries, which could be useful for identifying a specific subset of cards but does not provide deeper insights or trends.
Filtered Qwen 7B Model Cards
The query filters for specific terms related to "distilled" or "distill", "qwen", and "7b" in the 'card' column but excludes certain base models, providing a limited set of entries for further inspection.
Qwen 7B Distilled Models
The query provides a basic filtering of records to find specific card names that include keywords related to distilled Qwen 7b models, excluding a particular base model, which gives limited insight but helps in focusing on relevant entries.
Qwen 7B Distilled Model Cards
The query filters data based on specific keywords in the modelId and card fields, providing limited insight primarily useful for locating specific entries rather than revealing broad patterns or trends.
Qwen 7B Distilled Models
Finds all entries containing the terms 'distilled', 'qwen', and '7b' in a case-insensitive manner, providing a filtered set of records but without deeper analysis.
Distilled Qwen 7B Models
The query filters for specific model IDs containing 'distilled', 'qwen', and '7b', providing a basic retrieval of relevant entries but without deeper analysis or insight.
Filtered Model Cards with Distill Qwen2.
Filters and retrieves records containing specific keywords in the card description while excluding certain phrases, providing a basic count of relevant entries.
Filtered Model Cards with Distill Qwen 7
The query filters specific variations of card descriptions containing 'distill', 'qwen', and '7b' while excluding a particular base model, providing limited but specific data retrieval.
Distill Qwen 7B Model Cards
The query filters and retrieves rows where the 'card' column contains specific keywords ('distill', 'qwen', and '7b'), providing a basic filter result that can help in identifying specific entries.