metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: TheBloke/Mistral-7B-Instruct-v0.2-GPTQ
model-index:
- name: aiHumangpt-ft
results: []
aiHumangpt-ft
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: 2.5504
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: 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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.058 | 0.91 | 5 | 2.7601 |
| 2.3878 | 2.0 | 11 | 2.6707 |
| 2.7956 | 2.91 | 16 | 2.6278 |
| 2.2611 | 4.0 | 22 | 2.6003 |
| 2.6933 | 4.91 | 27 | 2.5844 |
| 2.2262 | 6.0 | 33 | 2.5682 |
| 2.63 | 6.91 | 38 | 2.5592 |
| 2.1795 | 8.0 | 44 | 2.5528 |
| 2.5948 | 8.91 | 49 | 2.5505 |
| 1.2205 | 9.09 | 50 | 2.5504 |
Framework versions
- PEFT 0.10.0
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2