metadata
license: other
library_name: peft
tags:
- generated_from_trainer
base_model: TheBloke/Wizard-Vicuna-7B-Uncensored-GPTQ
model-index:
- name: amadeus_v1
results: []
amadeus_v1
This model is a fine-tuned version of TheBloke/Wizard-Vicuna-7B-Uncensored-GPTQ on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 4.5349
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: 0.002
- 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: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.4924 | 1.0 | 75 | 3.2433 |
| 3.1246 | 2.0 | 150 | 3.2148 |
| 2.8655 | 3.0 | 225 | 3.2529 |
| 2.6201 | 4.0 | 300 | 3.3286 |
| 2.4044 | 5.0 | 375 | 3.4331 |
| 2.2106 | 6.0 | 450 | 3.5828 |
| 2.0348 | 7.0 | 525 | 3.7128 |
| 1.9341 | 8.0 | 600 | 3.7719 |
| 1.7997 | 9.0 | 675 | 3.8238 |
| 1.7071 | 10.0 | 750 | 3.9716 |
| 1.6537 | 11.0 | 825 | 4.0883 |
| 1.5708 | 12.0 | 900 | 4.1066 |
| 1.5302 | 13.0 | 975 | 4.1778 |
| 1.4895 | 14.0 | 1050 | 4.2198 |
| 1.4586 | 15.0 | 1125 | 4.2890 |
| 1.4211 | 16.0 | 1200 | 4.3352 |
| 1.3721 | 17.0 | 1275 | 4.4004 |
| 1.3637 | 18.0 | 1350 | 4.4707 |
| 1.3597 | 19.0 | 1425 | 4.5228 |
| 1.3362 | 20.0 | 1500 | 4.5349 |
Framework versions
- PEFT 0.10.0
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2