metadata
license: apache-2.0
library_name: peft
tags:
- alignment-handbook
- generated_from_trainer
- trl
- sft
base_model: mistralai/Mistral-7B-Instruct-v0.2
datasets:
- generator
model-index:
- name: data
results: []
data
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 1.2003
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: 0.0002
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1
- training_steps: 1000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.9474 | 0.1479 | 25 | 0.6950 |
| 0.6403 | 0.2959 | 50 | 0.6291 |
| 0.6184 | 0.4438 | 75 | 0.6191 |
| 0.6041 | 0.5917 | 100 | 0.6083 |
| 0.6023 | 0.7396 | 125 | 0.6047 |
| 0.6086 | 0.8876 | 150 | 0.6014 |
| 0.5545 | 1.0355 | 175 | 0.6375 |
| 0.4278 | 1.1834 | 200 | 0.6235 |
| 0.4297 | 1.3314 | 225 | 0.6276 |
| 0.4172 | 1.4793 | 250 | 0.6314 |
| 0.4273 | 1.6272 | 275 | 0.6375 |
| 0.4264 | 1.7751 | 300 | 0.6345 |
| 0.428 | 1.9231 | 325 | 0.6296 |
| 0.3524 | 2.0710 | 350 | 0.7067 |
| 0.2697 | 2.2189 | 375 | 0.7314 |
| 0.2645 | 2.3669 | 400 | 0.7255 |
| 0.2778 | 2.5148 | 425 | 0.7221 |
| 0.2687 | 2.6627 | 450 | 0.7373 |
| 0.2748 | 2.8107 | 475 | 0.7250 |
| 0.2737 | 2.9586 | 500 | 0.7214 |
| 0.1848 | 3.1065 | 525 | 0.8414 |
| 0.1535 | 3.2544 | 550 | 0.8438 |
| 0.1565 | 3.4024 | 575 | 0.8479 |
| 0.1583 | 3.5503 | 600 | 0.8719 |
| 0.1537 | 3.6982 | 625 | 0.8497 |
| 0.1611 | 3.8462 | 650 | 0.8587 |
| 0.16 | 3.9941 | 675 | 0.8714 |
| 0.0913 | 4.1420 | 700 | 1.0194 |
| 0.0841 | 4.2899 | 725 | 1.0429 |
| 0.0836 | 4.4379 | 750 | 1.0544 |
| 0.0842 | 4.5858 | 775 | 1.0432 |
| 0.081 | 4.7337 | 800 | 1.0386 |
| 0.0848 | 4.8817 | 825 | 1.0705 |
| 0.0776 | 5.0296 | 850 | 1.0773 |
| 0.0485 | 5.1775 | 875 | 1.1910 |
| 0.0481 | 5.3254 | 900 | 1.1800 |
| 0.0481 | 5.4734 | 925 | 1.1915 |
| 0.0467 | 5.6213 | 950 | 1.1996 |
| 0.0468 | 5.7692 | 975 | 1.1961 |
| 0.0449 | 5.9172 | 1000 | 1.2003 |
Framework versions
- PEFT 0.10.0
- Transformers 4.40.0
- Pytorch 2.2.2
- Datasets 2.19.0
- Tokenizers 0.19.1