|
|
--- |
|
|
base_model: meta-llama/Llama-2-7b-hf |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: qlora-out |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
|
|
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl) |
|
|
# qlora-out |
|
|
|
|
|
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.5703 |
|
|
|
|
|
## 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: 0.0004 |
|
|
- train_batch_size: 1 |
|
|
- eval_batch_size: 1 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 16 |
|
|
- total_train_batch_size: 16 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: cosine |
|
|
- lr_scheduler_warmup_steps: 100 |
|
|
- num_epochs: 2 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:----:|:---------------:| |
|
|
| 0.8756 | 0.06 | 20 | 0.7111 | |
|
|
| 0.9058 | 0.11 | 40 | 0.6764 | |
|
|
| 0.7526 | 0.17 | 60 | 0.6669 | |
|
|
| 0.6926 | 0.23 | 80 | 0.6363 | |
|
|
| 0.6731 | 0.28 | 100 | 0.6187 | |
|
|
| 0.647 | 0.34 | 120 | 0.6162 | |
|
|
| 0.6219 | 0.4 | 140 | 0.6041 | |
|
|
| 0.5781 | 0.45 | 160 | 0.5937 | |
|
|
| 0.6346 | 0.51 | 180 | 0.6006 | |
|
|
| 0.7663 | 0.57 | 200 | 0.5926 | |
|
|
| 0.5864 | 0.62 | 220 | 0.5866 | |
|
|
| 0.5943 | 0.68 | 240 | 0.5756 | |
|
|
| 0.5029 | 0.74 | 260 | 0.5733 | |
|
|
| 0.5482 | 0.79 | 280 | 0.5712 | |
|
|
| 0.5413 | 0.85 | 300 | 0.5820 | |
|
|
| 0.657 | 0.91 | 320 | 0.5696 | |
|
|
| 0.506 | 0.96 | 340 | 0.5839 | |
|
|
| 0.4804 | 1.02 | 360 | 0.5803 | |
|
|
| 0.5095 | 1.08 | 380 | 0.5974 | |
|
|
| 0.4404 | 1.13 | 400 | 0.5746 | |
|
|
| 0.3869 | 1.19 | 420 | 0.5740 | |
|
|
| 0.4129 | 1.25 | 440 | 0.5777 | |
|
|
| 0.4209 | 1.3 | 460 | 0.5825 | |
|
|
| 0.4014 | 1.36 | 480 | 0.5742 | |
|
|
| 0.3333 | 1.42 | 500 | 0.5851 | |
|
|
| 0.5041 | 1.47 | 520 | 0.5798 | |
|
|
| 0.5528 | 1.53 | 540 | 0.5631 | |
|
|
| 0.4372 | 1.59 | 560 | 0.5747 | |
|
|
| 0.3901 | 1.64 | 580 | 0.5625 | |
|
|
| 0.5271 | 1.7 | 600 | 0.5746 | |
|
|
| 0.4283 | 1.76 | 620 | 0.5662 | |
|
|
| 0.4336 | 1.81 | 640 | 0.5652 | |
|
|
| 0.3534 | 1.87 | 660 | 0.5697 | |
|
|
| 0.4728 | 1.93 | 680 | 0.5713 | |
|
|
| 0.5159 | 1.98 | 700 | 0.5703 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.34.1 |
|
|
- Pytorch 2.0.1+cu118 |
|
|
- Datasets 2.14.6 |
|
|
- Tokenizers 0.14.1 |
|
|
|
