|
--- |
|
base_model: meta-llama/Llama-2-13b-hf |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: radiopaedia_cl-llama2_13b-240311 |
|
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. --> |
|
|
|
# radiopaedia_cl-llama2_13b-240311 |
|
|
|
This model is a fine-tuned version of [meta-llama/Llama-2-13b-hf](https://huggingface.co./meta-llama/Llama-2-13b-hf) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.7476 |
|
|
|
## 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: 0.0003 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 16 |
|
- 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: 2 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 0.8773 | 0.05 | 20 | 0.8701 | |
|
| 0.8515 | 0.11 | 40 | 0.8439 | |
|
| 0.8288 | 0.16 | 60 | 0.8311 | |
|
| 0.8047 | 0.21 | 80 | 0.8179 | |
|
| 0.7462 | 0.27 | 100 | 0.8031 | |
|
| 0.852 | 0.32 | 120 | 0.7854 | |
|
| 0.8301 | 0.37 | 140 | 0.7937 | |
|
| 0.8407 | 0.42 | 160 | 0.7798 | |
|
| 0.7099 | 0.48 | 180 | 0.7690 | |
|
| 0.8138 | 0.53 | 200 | 0.7630 | |
|
| 0.8015 | 0.58 | 220 | 0.7670 | |
|
| 0.6831 | 0.64 | 240 | 0.7592 | |
|
| 0.7111 | 0.69 | 260 | 0.7601 | |
|
| 0.6946 | 0.74 | 280 | 0.7461 | |
|
| 0.6754 | 0.8 | 300 | 0.7514 | |
|
| 0.8267 | 0.85 | 320 | 0.7388 | |
|
| 0.7297 | 0.9 | 340 | 0.7372 | |
|
| 0.6722 | 0.96 | 360 | 0.7306 | |
|
| 0.5094 | 1.01 | 380 | 0.7278 | |
|
| 0.49 | 1.06 | 400 | 0.7530 | |
|
| 0.5015 | 1.12 | 420 | 0.7643 | |
|
| 0.4161 | 1.17 | 440 | 0.7585 | |
|
| 0.5348 | 1.22 | 460 | 0.7661 | |
|
| 0.4794 | 1.27 | 480 | 0.7525 | |
|
| 0.4915 | 1.33 | 500 | 0.7573 | |
|
| 0.4564 | 1.38 | 520 | 0.7667 | |
|
| 0.5295 | 1.43 | 540 | 0.7640 | |
|
| 0.503 | 1.49 | 560 | 0.7604 | |
|
| 0.3933 | 1.54 | 580 | 0.7602 | |
|
| 0.4384 | 1.59 | 600 | 0.7655 | |
|
| 0.406 | 1.65 | 620 | 0.7531 | |
|
| 0.5174 | 1.7 | 640 | 0.7473 | |
|
| 0.4811 | 1.75 | 660 | 0.7457 | |
|
| 0.4586 | 1.81 | 680 | 0.7519 | |
|
| 0.4957 | 1.86 | 700 | 0.7469 | |
|
| 0.4006 | 1.91 | 720 | 0.7458 | |
|
| 0.454 | 1.97 | 740 | 0.7476 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.0.1+cu117 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
