metadata
base_model: TheBloke/Mistral-7B-Instruct-v0.2-GPTQ
library_name: peft
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: shawgpt-ft-lr2e-05-wd0.01
results: []
shawgpt-ft-lr2e-05-wd0.01
This model is a fine-tuned version of TheBloke/Mistral-7B-Instruct-v0.2-GPTQ on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.9690
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- 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: 2
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.6447 | 0.9231 | 3 | 4.2183 |
| 4.5908 | 1.8462 | 6 | 4.1843 |
| 4.5642 | 2.7692 | 9 | 4.1518 |
| 3.3876 | 4.0 | 13 | 4.1100 |
| 4.4932 | 4.9231 | 16 | 4.0795 |
| 4.4358 | 5.8462 | 19 | 4.0517 |
| 4.4138 | 6.7692 | 22 | 4.0268 |
| 3.2645 | 8.0 | 26 | 3.9996 |
| 4.335 | 8.9231 | 29 | 3.9846 |
| 4.3323 | 9.8462 | 32 | 3.9746 |
| 4.3194 | 10.7692 | 35 | 3.9696 |
| 1.0224 | 11.0769 | 36 | 3.9690 |
Framework versions
- PEFT 0.13.2
- Transformers 4.44.2
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.19.1