|
--- |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
base_model: NousResearch/Llama-2-7b-hf |
|
model-index: |
|
- name: results |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# results |
|
|
|
This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.9343 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 16 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 10 |
|
- num_epochs: 4 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.7949 | 0.1 | 25 | 1.7390 | |
|
| 1.7268 | 0.21 | 50 | 1.6898 | |
|
| 1.7214 | 0.31 | 75 | 1.6547 | |
|
| 1.6573 | 0.41 | 100 | 1.6260 | |
|
| 1.6663 | 0.52 | 125 | 1.5999 | |
|
| 1.6499 | 0.62 | 150 | 1.5735 | |
|
| 1.61 | 0.72 | 175 | 1.5483 | |
|
| 1.6491 | 0.82 | 200 | 1.5288 | |
|
| 1.6441 | 0.93 | 225 | 1.5034 | |
|
| 1.5055 | 1.03 | 250 | 1.4753 | |
|
| 1.4955 | 1.13 | 275 | 1.4409 | |
|
| 1.4345 | 1.24 | 300 | 1.4159 | |
|
| 1.449 | 1.34 | 325 | 1.3918 | |
|
| 1.4762 | 1.44 | 350 | 1.3701 | |
|
| 1.4562 | 1.55 | 375 | 1.3449 | |
|
| 1.4092 | 1.65 | 400 | 1.3205 | |
|
| 1.4173 | 1.75 | 425 | 1.3000 | |
|
| 1.4225 | 1.86 | 450 | 1.2798 | |
|
| 1.4358 | 1.96 | 475 | 1.2600 | |
|
| 1.2387 | 2.06 | 500 | 1.2198 | |
|
| 1.232 | 2.16 | 525 | 1.1936 | |
|
| 1.2212 | 2.27 | 550 | 1.1720 | |
|
| 1.2066 | 2.37 | 575 | 1.1491 | |
|
| 1.199 | 2.47 | 600 | 1.1301 | |
|
| 1.2365 | 2.58 | 625 | 1.1124 | |
|
| 1.2124 | 2.68 | 650 | 1.0934 | |
|
| 1.2113 | 2.78 | 675 | 1.0766 | |
|
| 1.1947 | 2.89 | 700 | 1.0620 | |
|
| 1.2041 | 2.99 | 725 | 1.0441 | |
|
| 1.0437 | 3.09 | 750 | 1.0133 | |
|
| 1.0239 | 3.2 | 775 | 0.9972 | |
|
| 1.0417 | 3.3 | 800 | 0.9839 | |
|
| 0.9909 | 3.4 | 825 | 0.9712 | |
|
| 1.0275 | 3.51 | 850 | 0.9600 | |
|
| 1.0466 | 3.61 | 875 | 0.9516 | |
|
| 1.0036 | 3.71 | 900 | 0.9423 | |
|
| 1.0039 | 3.81 | 925 | 0.9363 | |
|
| 0.9786 | 3.92 | 950 | 0.9343 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.8.2 |
|
- Transformers 4.37.2 |
|
- Pytorch 2.2.0+cu121 |
|
- Datasets 2.17.0 |
|
- Tokenizers 0.15.1 |
