|
|
--- |
|
|
library_name: transformers |
|
|
tags: |
|
|
- trl |
|
|
- sft |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: llama_1b_step2_batch_v4 |
|
|
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_v4 |
|
|
|
|
|
This model was trained from scratch on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.3246 |
|
|
|
|
|
## 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: 4 |
|
|
- total_train_batch_size: 32 |
|
|
- 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 | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 0.7442 | 0.1363 | 50 | 0.8824 | |
|
|
| 0.5925 | 0.2727 | 100 | 0.6583 | |
|
|
| 0.587 | 0.4090 | 150 | 0.5422 | |
|
|
| 0.4862 | 0.5453 | 200 | 0.4668 | |
|
|
| 0.4211 | 0.6817 | 250 | 0.4176 | |
|
|
| 0.4013 | 0.8180 | 300 | 0.3846 | |
|
|
| 0.4198 | 0.9543 | 350 | 0.3644 | |
|
|
| 0.2427 | 1.0913 | 400 | 0.3562 | |
|
|
| 0.2783 | 1.2277 | 450 | 0.3468 | |
|
|
| 0.22 | 1.3640 | 500 | 0.3399 | |
|
|
| 0.2263 | 1.5003 | 550 | 0.3335 | |
|
|
| 0.2177 | 1.6367 | 600 | 0.3277 | |
|
|
| 0.2579 | 1.7730 | 650 | 0.3251 | |
|
|
| 0.1985 | 1.9093 | 700 | 0.3246 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.46.0 |
|
|
- Pytorch 2.1.0+cu118 |
|
|
- Datasets 3.0.2 |
|
|
- Tokenizers 0.20.1 |
|
|
|
