|
--- |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
base_model: NousResearch/Llama-2-7b-hf |
|
model-index: |
|
- name: llama2_instruct_generation |
|
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. --> |
|
|
|
# llama2_instruct_generation |
|
|
|
This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on the generator dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.6908 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 0.0002 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: constant |
|
- lr_scheduler_warmup_steps: 0.03 |
|
- training_steps: 500 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.7099 | 0.0 | 20 | 1.8211 | |
|
| 1.9047 | 0.0 | 40 | 1.7987 | |
|
| 1.9396 | 0.0 | 60 | 1.7830 | |
|
| 1.9942 | 0.0 | 80 | 1.7758 | |
|
| 1.8507 | 0.0 | 100 | 1.7698 | |
|
| 2.0794 | 0.0 | 120 | 1.7683 | |
|
| 1.8043 | 0.0 | 140 | 1.7646 | |
|
| 2.0321 | 0.01 | 160 | 1.7563 | |
|
| 1.7889 | 0.01 | 180 | 1.7487 | |
|
| 1.8727 | 0.01 | 200 | 1.7279 | |
|
| 1.6872 | 0.01 | 220 | 1.7131 | |
|
| 1.6098 | 0.01 | 240 | 1.7100 | |
|
| 2.0663 | 0.01 | 260 | 1.7101 | |
|
| 1.8513 | 0.01 | 280 | 1.7054 | |
|
| 1.7695 | 0.01 | 300 | 1.7015 | |
|
| 1.7101 | 0.01 | 320 | 1.7015 | |
|
| 1.7896 | 0.01 | 340 | 1.6957 | |
|
| 2.0864 | 0.01 | 360 | 1.6974 | |
|
| 1.8656 | 0.01 | 380 | 1.6949 | |
|
| 1.8073 | 0.01 | 400 | 1.6928 | |
|
| 1.7276 | 0.01 | 420 | 1.6911 | |
|
| 1.7795 | 0.01 | 440 | 1.6897 | |
|
| 1.8992 | 0.02 | 460 | 1.6882 | |
|
| 1.5421 | 0.02 | 480 | 1.6889 | |
|
| 1.9574 | 0.02 | 500 | 1.6908 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.7.1 |
|
- Transformers 4.37.0 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
