metadata
library_name: peft
license: llama3.1
base_model: meta-llama/Llama-3.1-8B-Instruct
tags:
- llama-factory
- lora
- generated_from_trainer
model-index:
- name: Llama-3.1-8B-Instruct-PsyCourse-fold7
results: []
Llama-3.1-8B-Instruct-PsyCourse-fold7
This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on the course-train-fold7 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0333
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: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.9681 | 0.0764 | 50 | 0.6998 |
| 0.2016 | 0.1528 | 100 | 0.1420 |
| 0.0868 | 0.2292 | 150 | 0.0756 |
| 0.0662 | 0.3056 | 200 | 0.0589 |
| 0.0623 | 0.3820 | 250 | 0.0552 |
| 0.0458 | 0.4584 | 300 | 0.0502 |
| 0.0681 | 0.5348 | 350 | 0.0517 |
| 0.0451 | 0.6112 | 400 | 0.0472 |
| 0.0596 | 0.6875 | 450 | 0.0469 |
| 0.0478 | 0.7639 | 500 | 0.0419 |
| 0.0329 | 0.8403 | 550 | 0.0406 |
| 0.0545 | 0.9167 | 600 | 0.0410 |
| 0.0586 | 0.9931 | 650 | 0.0452 |
| 0.0407 | 1.0695 | 700 | 0.0391 |
| 0.029 | 1.1459 | 750 | 0.0369 |
| 0.0345 | 1.2223 | 800 | 0.0397 |
| 0.0399 | 1.2987 | 850 | 0.0395 |
| 0.0419 | 1.3751 | 900 | 0.0393 |
| 0.0482 | 1.4515 | 950 | 0.0405 |
| 0.0329 | 1.5279 | 1000 | 0.0361 |
| 0.0306 | 1.6043 | 1050 | 0.0381 |
| 0.0308 | 1.6807 | 1100 | 0.0385 |
| 0.0612 | 1.7571 | 1150 | 0.0365 |
| 0.0369 | 1.8335 | 1200 | 0.0347 |
| 0.0394 | 1.9099 | 1250 | 0.0394 |
| 0.0325 | 1.9862 | 1300 | 0.0373 |
| 0.0267 | 2.0626 | 1350 | 0.0364 |
| 0.0236 | 2.1390 | 1400 | 0.0353 |
| 0.0178 | 2.2154 | 1450 | 0.0401 |
| 0.0261 | 2.2918 | 1500 | 0.0350 |
| 0.024 | 2.3682 | 1550 | 0.0350 |
| 0.0215 | 2.4446 | 1600 | 0.0339 |
| 0.0316 | 2.5210 | 1650 | 0.0384 |
| 0.0264 | 2.5974 | 1700 | 0.0362 |
| 0.027 | 2.6738 | 1750 | 0.0379 |
| 0.0366 | 2.7502 | 1800 | 0.0333 |
| 0.0303 | 2.8266 | 1850 | 0.0336 |
| 0.0264 | 2.9030 | 1900 | 0.0353 |
| 0.0213 | 2.9794 | 1950 | 0.0371 |
| 0.0188 | 3.0558 | 2000 | 0.0368 |
| 0.0199 | 3.1322 | 2050 | 0.0359 |
| 0.0149 | 3.2086 | 2100 | 0.0406 |
| 0.0179 | 3.2850 | 2150 | 0.0434 |
| 0.0209 | 3.3613 | 2200 | 0.0373 |
| 0.0199 | 3.4377 | 2250 | 0.0443 |
| 0.0169 | 3.5141 | 2300 | 0.0365 |
| 0.024 | 3.5905 | 2350 | 0.0377 |
| 0.0182 | 3.6669 | 2400 | 0.0404 |
| 0.0138 | 3.7433 | 2450 | 0.0410 |
| 0.0197 | 3.8197 | 2500 | 0.0382 |
| 0.0201 | 3.8961 | 2550 | 0.0362 |
| 0.0129 | 3.9725 | 2600 | 0.0420 |
| 0.0084 | 4.0489 | 2650 | 0.0419 |
| 0.01 | 4.1253 | 2700 | 0.0444 |
| 0.009 | 4.2017 | 2750 | 0.0554 |
| 0.0075 | 4.2781 | 2800 | 0.0449 |
| 0.0123 | 4.3545 | 2850 | 0.0445 |
| 0.0086 | 4.4309 | 2900 | 0.0446 |
| 0.0128 | 4.5073 | 2950 | 0.0410 |
| 0.011 | 4.5837 | 3000 | 0.0446 |
| 0.0079 | 4.6600 | 3050 | 0.0467 |
| 0.0063 | 4.7364 | 3100 | 0.0447 |
| 0.0081 | 4.8128 | 3150 | 0.0446 |
| 0.0092 | 4.8892 | 3200 | 0.0423 |
| 0.0105 | 4.9656 | 3250 | 0.0434 |
| 0.0049 | 5.0420 | 3300 | 0.0503 |
| 0.006 | 5.1184 | 3350 | 0.0521 |
| 0.0037 | 5.1948 | 3400 | 0.0545 |
| 0.0032 | 5.2712 | 3450 | 0.0743 |
| 0.0047 | 5.3476 | 3500 | 0.0558 |
| 0.0037 | 5.4240 | 3550 | 0.0517 |
| 0.0054 | 5.5004 | 3600 | 0.0526 |
| 0.0053 | 5.5768 | 3650 | 0.0507 |
| 0.0094 | 5.6532 | 3700 | 0.0504 |
| 0.0067 | 5.7296 | 3750 | 0.0492 |
| 0.0038 | 5.8060 | 3800 | 0.0524 |
| 0.0091 | 5.8824 | 3850 | 0.0443 |
| 0.006 | 5.9587 | 3900 | 0.0490 |
| 0.0042 | 6.0351 | 3950 | 0.0518 |
| 0.0023 | 6.1115 | 4000 | 0.0607 |
| 0.0012 | 6.1879 | 4050 | 0.0625 |
| 0.0046 | 6.2643 | 4100 | 0.0562 |
| 0.0017 | 6.3407 | 4150 | 0.0639 |
| 0.0029 | 6.4171 | 4200 | 0.0585 |
| 0.0023 | 6.4935 | 4250 | 0.0586 |
| 0.0021 | 6.5699 | 4300 | 0.0601 |
| 0.0004 | 6.6463 | 4350 | 0.0675 |
| 0.0021 | 6.7227 | 4400 | 0.0667 |
| 0.0024 | 6.7991 | 4450 | 0.0701 |
| 0.0022 | 6.8755 | 4500 | 0.0674 |
| 0.0033 | 6.9519 | 4550 | 0.0609 |
| 0.0009 | 7.0283 | 4600 | 0.0551 |
| 0.0011 | 7.1047 | 4650 | 0.0607 |
| 0.0014 | 7.1811 | 4700 | 0.0657 |
| 0.0003 | 7.2574 | 4750 | 0.0645 |
| 0.0013 | 7.3338 | 4800 | 0.0692 |
| 0.0004 | 7.4102 | 4850 | 0.0737 |
| 0.0004 | 7.4866 | 4900 | 0.0669 |
| 0.0028 | 7.5630 | 4950 | 0.0651 |
| 0.0008 | 7.6394 | 5000 | 0.0633 |
| 0.0014 | 7.7158 | 5050 | 0.0643 |
| 0.0012 | 7.7922 | 5100 | 0.0659 |
| 0.0006 | 7.8686 | 5150 | 0.0663 |
| 0.0005 | 7.9450 | 5200 | 0.0700 |
| 0.0007 | 8.0214 | 5250 | 0.0659 |
| 0.0009 | 8.0978 | 5300 | 0.0691 |
| 0.0002 | 8.1742 | 5350 | 0.0709 |
| 0.0004 | 8.2506 | 5400 | 0.0735 |
| 0.0006 | 8.3270 | 5450 | 0.0750 |
| 0.0007 | 8.4034 | 5500 | 0.0772 |
| 0.0001 | 8.4798 | 5550 | 0.0785 |
| 0.0007 | 8.5561 | 5600 | 0.0807 |
| 0.0013 | 8.6325 | 5650 | 0.0787 |
| 0.0005 | 8.7089 | 5700 | 0.0770 |
| 0.0012 | 8.7853 | 5750 | 0.0768 |
| 0.0004 | 8.8617 | 5800 | 0.0756 |
| 0.0032 | 8.9381 | 5850 | 0.0763 |
| 0.0008 | 9.0145 | 5900 | 0.0764 |
| 0.0002 | 9.0909 | 5950 | 0.0777 |
| 0.0002 | 9.1673 | 6000 | 0.0781 |
| 0.0003 | 9.2437 | 6050 | 0.0786 |
| 0.0007 | 9.3201 | 6100 | 0.0790 |
| 0.0002 | 9.3965 | 6150 | 0.0798 |
| 0.0003 | 9.4729 | 6200 | 0.0796 |
| 0.0008 | 9.5493 | 6250 | 0.0798 |
| 0.0002 | 9.6257 | 6300 | 0.0800 |
| 0.0012 | 9.7021 | 6350 | 0.0801 |
| 0.0008 | 9.7785 | 6400 | 0.0801 |
| 0.0001 | 9.8549 | 6450 | 0.0801 |
| 0.0005 | 9.9312 | 6500 | 0.0802 |
Framework versions
- PEFT 0.12.0
- Transformers 4.46.1
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3