metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: TheBloke/Mistral-7B-Instruct-v0.2-GPTQ
model-index:
- name: mistral-triples-ft
results: []
mistral-triples-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: 1.3557
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 |
|---|---|---|---|
| 4.5009 | 1.0 | 6 | 3.7858 |
| 3.3363 | 2.0 | 12 | 2.7896 |
| 2.5177 | 3.0 | 18 | 2.0983 |
| 1.9915 | 4.0 | 24 | 1.7282 |
| 1.7126 | 5.0 | 30 | 1.5479 |
| 1.5714 | 6.0 | 36 | 1.4878 |
| 1.5257 | 7.0 | 42 | 1.4403 |
| 1.4793 | 8.0 | 48 | 1.4035 |
| 1.4395 | 9.0 | 54 | 1.3699 |
| 1.417 | 10.0 | 60 | 1.3557 |
Framework versions
- PEFT 0.10.0
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2