metadata
base_model: mistralai/Mistral-7B-Instruct-v0.1
datasets:
- generator
library_name: peft
license: apache-2.0
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: mistral_recipe1m_5500
results: []
mistral_recipe1m_5500
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: 1.1857
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_steps: 3
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.3671 | 0.1081 | 20 | 1.3132 |
| 1.2542 | 0.2162 | 40 | 1.2499 |
| 1.2247 | 0.3243 | 60 | 1.2297 |
| 1.2297 | 0.4324 | 80 | 1.2183 |
| 1.2271 | 0.5405 | 100 | 1.2078 |
| 1.2063 | 0.6486 | 120 | 1.2005 |
| 1.2123 | 0.7568 | 140 | 1.1947 |
| 1.1968 | 0.8649 | 160 | 1.1903 |
| 1.2038 | 0.9730 | 180 | 1.1857 |
Framework versions
- PEFT 0.13.2
- Transformers 4.45.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0