metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
datasets:
- generator
base_model: mistralai/Mixtral-8x7B-Instruct-v0.1
model-index:
- name: Mixtral_R2_v0
results: []
Mixtral_R2_v0
This model is a fine-tuned version of mistralai/Mixtral-8x7B-Instruct-v0.1 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.7382
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: 2.5e-05
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.03
- num_epochs: 25
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.2053 | 1.0 | 18 | 1.6050 |
| 1.685 | 2.0 | 36 | 1.4131 |
| 1.4913 | 3.0 | 54 | 1.2708 |
| 1.3079 | 4.0 | 72 | 1.1353 |
| 1.1686 | 5.0 | 90 | 1.0360 |
| 1.1356 | 6.0 | 108 | 0.9803 |
| 1.053 | 7.0 | 126 | 0.9382 |
| 0.9968 | 8.0 | 144 | 0.8994 |
| 0.947 | 9.0 | 162 | 0.8616 |
| 0.8981 | 10.0 | 180 | 0.8244 |
| 0.8726 | 11.0 | 198 | 0.7891 |
| 0.8211 | 12.0 | 216 | 0.7727 |
| 0.8126 | 13.0 | 234 | 0.7645 |
| 0.7889 | 14.0 | 252 | 0.7591 |
| 0.7837 | 15.0 | 270 | 0.7547 |
| 0.7792 | 16.0 | 288 | 0.7510 |
| 0.7507 | 17.0 | 306 | 0.7481 |
| 0.7483 | 18.0 | 324 | 0.7458 |
| 0.7429 | 19.0 | 342 | 0.7432 |
| 0.7362 | 20.0 | 360 | 0.7417 |
| 0.7495 | 21.0 | 378 | 0.7406 |
| 0.748 | 22.0 | 396 | 0.7397 |
| 0.7429 | 23.0 | 414 | 0.7388 |
| 0.7444 | 24.0 | 432 | 0.7384 |
| 0.7334 | 25.0 | 450 | 0.7382 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0