metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
datasets:
- generator
base_model: mistralai/Mistral-7B-Instruct-v0.1
model-index:
- name: mistral_7b_instruct_v0_1_txt_2_sql
results: []
mistral_7b_instruct_v0_1_txt_2_sql
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.6482
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_steps: 0.03
- training_steps: 80
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.3599 | 0.71 | 10 | 1.0764 |
| 1.0005 | 1.43 | 20 | 0.8726 |
| 0.8508 | 2.14 | 30 | 0.7721 |
| 0.7585 | 2.86 | 40 | 0.7218 |
| 0.7078 | 3.57 | 50 | 0.6825 |
| 0.6413 | 4.29 | 60 | 0.6594 |
| 0.6128 | 5.0 | 70 | 0.6508 |
| 0.5745 | 5.71 | 80 | 0.6482 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0