|
|
--- |
|
|
library_name: peft |
|
|
license: llama3 |
|
|
base_model: aaditya/Llama3-OpenBioLLM-8B |
|
|
tags: |
|
|
- llama-factory |
|
|
- lora |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: Llama3-OpenBioLLM-8B-PsyCourse-fold8 |
|
|
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. --> |
|
|
|
|
|
# Llama3-OpenBioLLM-8B-PsyCourse-fold8 |
|
|
|
|
|
This model is a fine-tuned version of [aaditya/Llama3-OpenBioLLM-8B](https://huggingface.co/aaditya/Llama3-OpenBioLLM-8B) on the course-train-fold8 dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.0360 |
|
|
|
|
|
## 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: 0.0001 |
|
|
- train_batch_size: 1 |
|
|
- eval_batch_size: 1 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 16 |
|
|
- 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: cosine |
|
|
- lr_scheduler_warmup_ratio: 0.1 |
|
|
- num_epochs: 5.0 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 0.5651 | 0.0758 | 50 | 0.3454 | |
|
|
| 0.159 | 0.1516 | 100 | 0.0916 | |
|
|
| 0.0775 | 0.2275 | 150 | 0.0660 | |
|
|
| 0.0565 | 0.3033 | 200 | 0.0587 | |
|
|
| 0.0563 | 0.3791 | 250 | 0.0594 | |
|
|
| 0.0631 | 0.4549 | 300 | 0.0575 | |
|
|
| 0.0677 | 0.5308 | 350 | 0.0503 | |
|
|
| 0.0366 | 0.6066 | 400 | 0.0471 | |
|
|
| 0.0397 | 0.6824 | 450 | 0.0430 | |
|
|
| 0.0383 | 0.7582 | 500 | 0.0479 | |
|
|
| 0.0508 | 0.8340 | 550 | 0.0427 | |
|
|
| 0.0346 | 0.9099 | 600 | 0.0434 | |
|
|
| 0.0513 | 0.9857 | 650 | 0.0444 | |
|
|
| 0.0339 | 1.0615 | 700 | 0.0417 | |
|
|
| 0.0296 | 1.1373 | 750 | 0.0442 | |
|
|
| 0.0288 | 1.2132 | 800 | 0.0397 | |
|
|
| 0.0299 | 1.2890 | 850 | 0.0421 | |
|
|
| 0.0293 | 1.3648 | 900 | 0.0401 | |
|
|
| 0.0278 | 1.4406 | 950 | 0.0393 | |
|
|
| 0.0283 | 1.5164 | 1000 | 0.0405 | |
|
|
| 0.0493 | 1.5923 | 1050 | 0.0393 | |
|
|
| 0.0287 | 1.6681 | 1100 | 0.0392 | |
|
|
| 0.0383 | 1.7439 | 1150 | 0.0379 | |
|
|
| 0.0312 | 1.8197 | 1200 | 0.0378 | |
|
|
| 0.0353 | 1.8956 | 1250 | 0.0379 | |
|
|
| 0.0242 | 1.9714 | 1300 | 0.0360 | |
|
|
| 0.0176 | 2.0472 | 1350 | 0.0413 | |
|
|
| 0.0132 | 2.1230 | 1400 | 0.0386 | |
|
|
| 0.0224 | 2.1988 | 1450 | 0.0413 | |
|
|
| 0.0198 | 2.2747 | 1500 | 0.0423 | |
|
|
| 0.0191 | 2.3505 | 1550 | 0.0429 | |
|
|
| 0.017 | 2.4263 | 1600 | 0.0412 | |
|
|
| 0.0194 | 2.5021 | 1650 | 0.0465 | |
|
|
| 0.0178 | 2.5780 | 1700 | 0.0439 | |
|
|
| 0.0238 | 2.6538 | 1750 | 0.0411 | |
|
|
| 0.0181 | 2.7296 | 1800 | 0.0414 | |
|
|
| 0.0128 | 2.8054 | 1850 | 0.0439 | |
|
|
| 0.0287 | 2.8812 | 1900 | 0.0410 | |
|
|
| 0.0202 | 2.9571 | 1950 | 0.0418 | |
|
|
| 0.011 | 3.0329 | 2000 | 0.0430 | |
|
|
| 0.005 | 3.1087 | 2050 | 0.0487 | |
|
|
| 0.0045 | 3.1845 | 2100 | 0.0502 | |
|
|
| 0.0072 | 3.2604 | 2150 | 0.0496 | |
|
|
| 0.0098 | 3.3362 | 2200 | 0.0482 | |
|
|
| 0.0089 | 3.4120 | 2250 | 0.0492 | |
|
|
| 0.0072 | 3.4878 | 2300 | 0.0486 | |
|
|
| 0.0116 | 3.5636 | 2350 | 0.0496 | |
|
|
| 0.0094 | 3.6395 | 2400 | 0.0489 | |
|
|
| 0.0055 | 3.7153 | 2450 | 0.0501 | |
|
|
| 0.0095 | 3.7911 | 2500 | 0.0529 | |
|
|
| 0.0113 | 3.8669 | 2550 | 0.0517 | |
|
|
| 0.0042 | 3.9428 | 2600 | 0.0518 | |
|
|
| 0.0021 | 4.0186 | 2650 | 0.0539 | |
|
|
| 0.0027 | 4.0944 | 2700 | 0.0573 | |
|
|
| 0.0017 | 4.1702 | 2750 | 0.0590 | |
|
|
| 0.0033 | 4.2460 | 2800 | 0.0603 | |
|
|
| 0.003 | 4.3219 | 2850 | 0.0618 | |
|
|
| 0.0013 | 4.3977 | 2900 | 0.0623 | |
|
|
| 0.003 | 4.4735 | 2950 | 0.0625 | |
|
|
| 0.0036 | 4.5493 | 3000 | 0.0631 | |
|
|
| 0.0017 | 4.6252 | 3050 | 0.0634 | |
|
|
| 0.0023 | 4.7010 | 3100 | 0.0635 | |
|
|
| 0.0028 | 4.7768 | 3150 | 0.0635 | |
|
|
| 0.0028 | 4.8526 | 3200 | 0.0637 | |
|
|
| 0.0021 | 4.9284 | 3250 | 0.0636 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.12.0 |
|
|
- Transformers 4.46.1 |
|
|
- Pytorch 2.5.1+cu124 |
|
|
- Datasets 3.1.0 |
|
|
- Tokenizers 0.20.3 |
