metadata
base_model: mistralai/Mistral-7B-Instruct-v0.3
datasets:
- generator
library_name: peft
license: apache-2.0
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: Mistral-7B-Text2SQL
results: []
Mistral-7B-Text2SQL
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.4643
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: 3
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 24
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- lr_scheduler_warmup_steps: 15
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.8346 | 0.4 | 10 | 0.7031 |
| 0.5882 | 0.8 | 20 | 0.5273 |
| 0.487 | 1.2 | 30 | 0.4850 |
| 0.4423 | 1.6 | 40 | 0.4675 |
| 0.4235 | 2.0 | 50 | 0.4564 |
| 0.3464 | 2.4 | 60 | 0.4690 |
| 0.3411 | 2.8 | 70 | 0.4643 |
Framework versions
- PEFT 0.13.2
- Transformers 4.45.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.1