|
--- |
|
base_model: meta-llama/Llama-2-7b-hf |
|
library_name: peft |
|
license: llama2 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: Llama-2-7b-hf_alpaca-clean_l0.0002_64 |
|
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. --> |
|
|
|
# Llama-2-7b-hf_alpaca-clean_l0.0002_64 |
|
|
|
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.6868 |
|
|
|
## 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: 1 |
|
- seed: 0 |
|
- gradient_accumulation_steps: 16 |
|
- total_train_batch_size: 16 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: constant |
|
- lr_scheduler_warmup_ratio: 0.03 |
|
- training_steps: 10000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 1.0346 | 0.0003 | 1 | 2.3927 | |
|
| 1.9904 | 0.0590 | 187 | 1.5667 | |
|
| 1.3058 | 0.1179 | 374 | 1.5624 | |
|
| 1.0583 | 0.1769 | 561 | 1.5607 | |
|
| 1.8361 | 0.2359 | 748 | 1.5273 | |
|
| 1.7038 | 0.2949 | 935 | 1.5075 | |
|
| 1.2223 | 0.3538 | 1122 | 1.5029 | |
|
| 1.0661 | 0.4128 | 1309 | 1.5042 | |
|
| 1.7942 | 0.4718 | 1496 | 1.4753 | |
|
| 1.5587 | 0.5307 | 1683 | 1.4771 | |
|
| 1.1685 | 0.5897 | 1870 | 1.4801 | |
|
| 1.0578 | 0.6487 | 2057 | 1.4936 | |
|
| 2.2233 | 0.7077 | 2244 | 1.4694 | |
|
| 1.6078 | 0.7666 | 2431 | 1.4656 | |
|
| 1.0611 | 0.8256 | 2618 | 1.4765 | |
|
| 0.9979 | 0.8846 | 2805 | 1.4737 | |
|
| 2.0584 | 0.9436 | 2992 | 1.4601 | |
|
| 0.9627 | 1.0025 | 3179 | 1.4579 | |
|
| 2.3813 | 1.0615 | 3366 | 1.4914 | |
|
| 1.2986 | 1.1205 | 3553 | 1.4847 | |
|
| 1.164 | 1.1794 | 3740 | 1.4857 | |
|
| 0.9181 | 1.2384 | 3927 | 1.4967 | |
|
| 1.7152 | 1.2974 | 4114 | 1.4807 | |
|
| 1.0827 | 1.3564 | 4301 | 1.4828 | |
|
| 0.9565 | 1.4153 | 4488 | 1.4858 | |
|
| 1.0499 | 1.4743 | 4675 | 1.4932 | |
|
| 1.838 | 1.5333 | 4862 | 1.4806 | |
|
| 1.3075 | 1.5922 | 5049 | 1.4770 | |
|
| 0.9147 | 1.6512 | 5236 | 1.4877 | |
|
| 0.9353 | 1.7102 | 5423 | 1.4889 | |
|
| 1.6525 | 1.7692 | 5610 | 1.4757 | |
|
| 1.0597 | 1.8281 | 5797 | 1.4812 | |
|
| 0.9384 | 1.8871 | 5984 | 1.4737 | |
|
| 2.0972 | 1.9461 | 6171 | 1.4749 | |
|
| 0.8698 | 2.0050 | 6358 | 1.4729 | |
|
| 0.8746 | 2.0640 | 6545 | 1.6185 | |
|
| 1.3664 | 2.1230 | 6732 | 1.5706 | |
|
| 0.8976 | 2.1820 | 6919 | 1.5368 | |
|
| 0.9512 | 2.2409 | 7106 | 1.5635 | |
|
| 0.957 | 2.2999 | 7293 | 1.6088 | |
|
| 1.1006 | 2.3589 | 7480 | 1.5504 | |
|
| 1.1033 | 2.4178 | 7667 | 1.5426 | |
|
| 0.9105 | 2.4768 | 7854 | 1.5907 | |
|
| 1.0444 | 2.5358 | 8041 | 1.5730 | |
|
| 1.3787 | 2.5948 | 8228 | 1.5404 | |
|
| 0.9126 | 2.6537 | 8415 | 1.5434 | |
|
| 0.8307 | 2.7127 | 8602 | 1.5716 | |
|
| 1.5571 | 2.7717 | 8789 | 1.5673 | |
|
| 1.1696 | 2.8307 | 8976 | 1.5473 | |
|
| 0.9802 | 2.8896 | 9163 | 1.5524 | |
|
| 0.8512 | 2.9486 | 9350 | 1.5740 | |
|
| 0.6861 | 3.0076 | 9537 | 1.5948 | |
|
| 0.8245 | 3.0665 | 9724 | 1.6846 | |
|
| 1.2366 | 3.1255 | 9911 | 1.6969 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.1.dev0 |
|
- Transformers 4.45.0.dev0 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.0 |
|
- Tokenizers 0.19.1 |
