|
|
--- |
|
|
library_name: transformers |
|
|
tags: |
|
|
- trl |
|
|
- sft |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: llama_1b_step2_batch_grad_v2 |
|
|
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_1b_step2_batch_grad_v2 |
|
|
|
|
|
This model was trained from scratch on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.3255 |
|
|
|
|
|
## 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: 40 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 2 |
|
|
- total_train_batch_size: 16 |
|
|
- optimizer: Use 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: 2 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 1.0959 | 0.0682 | 50 | 1.1193 | |
|
|
| 0.8353 | 0.1363 | 100 | 0.9093 | |
|
|
| 0.8185 | 0.2045 | 150 | 0.7860 | |
|
|
| 0.6476 | 0.2727 | 200 | 0.6876 | |
|
|
| 0.6211 | 0.3408 | 250 | 0.6156 | |
|
|
| 0.5487 | 0.4090 | 300 | 0.5546 | |
|
|
| 0.3495 | 0.4772 | 350 | 0.5092 | |
|
|
| 0.4872 | 0.5453 | 400 | 0.4798 | |
|
|
| 0.5069 | 0.6135 | 450 | 0.4531 | |
|
|
| 0.3867 | 0.6817 | 500 | 0.4269 | |
|
|
| 0.3495 | 0.7498 | 550 | 0.4099 | |
|
|
| 0.3642 | 0.8180 | 600 | 0.3949 | |
|
|
| 0.4026 | 0.8862 | 650 | 0.3778 | |
|
|
| 0.3184 | 0.9543 | 700 | 0.3638 | |
|
|
| 0.2109 | 1.0225 | 750 | 0.3620 | |
|
|
| 0.3049 | 1.0907 | 800 | 0.3598 | |
|
|
| 0.2303 | 1.1588 | 850 | 0.3548 | |
|
|
| 0.2837 | 1.2270 | 900 | 0.3498 | |
|
|
| 0.2812 | 1.2952 | 950 | 0.3418 | |
|
|
| 0.212 | 1.3633 | 1000 | 0.3414 | |
|
|
| 0.2756 | 1.4315 | 1050 | 0.3354 | |
|
|
| 0.2508 | 1.4997 | 1100 | 0.3329 | |
|
|
| 0.2125 | 1.5678 | 1150 | 0.3312 | |
|
|
| 0.1492 | 1.6360 | 1200 | 0.3294 | |
|
|
| 0.2738 | 1.7042 | 1250 | 0.3273 | |
|
|
| 0.2589 | 1.7723 | 1300 | 0.3264 | |
|
|
| 0.1845 | 1.8405 | 1350 | 0.3257 | |
|
|
| 0.2307 | 1.9087 | 1400 | 0.3255 | |
|
|
| 0.2417 | 1.9768 | 1450 | 0.3255 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.46.0 |
|
|
- Pytorch 2.1.0+cu118 |
|
|
- Datasets 3.0.2 |
|
|
- Tokenizers 0.20.1 |
|
|
|
