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_edit
results: []
shawgpt-ft_edit
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: 1.2911
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: 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: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.5538 | 0.9231 | 3 | 3.8017 |
| 3.8581 | 1.8462 | 6 | 3.2225 |
| 3.2748 | 2.7692 | 9 | 2.7911 |
| 2.1101 | 4.0 | 13 | 2.3250 |
| 2.3849 | 4.9231 | 16 | 2.0339 |
| 1.9759 | 5.8462 | 19 | 1.7558 |
| 1.6528 | 6.7692 | 22 | 1.5666 |
| 1.1169 | 8.0 | 26 | 1.4121 |
| 1.3772 | 8.9231 | 29 | 1.3694 |
| 1.3045 | 9.8462 | 32 | 1.3410 |
| 1.2927 | 10.7692 | 35 | 1.3183 |
| 0.917 | 12.0 | 39 | 1.2995 |
| 1.2326 | 12.9231 | 42 | 1.2932 |
| 1.0194 | 13.8462 | 45 | 1.2911 |
Framework versions
- PEFT 0.13.1
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1