metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: mistralai/Mistral-7B-Instruct-v0.2
model-index:
- name: mistral-7b-Javelin-small
results: []
mistral-7b-Javelin-small
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2137
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: linear
- training_steps: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.3516 | 0.07 | 1 | 2.0247 |
| 2.0225 | 0.14 | 2 | 1.8721 |
| 1.7928 | 0.21 | 3 | 1.8193 |
| 1.7108 | 0.29 | 4 | 1.7804 |
| 1.6954 | 0.36 | 5 | 1.7180 |
| 1.6487 | 0.43 | 6 | 1.6507 |
| 1.4139 | 0.5 | 7 | 1.5918 |
| 1.4606 | 0.57 | 8 | 1.5408 |
| 1.3855 | 0.64 | 9 | 1.5004 |
| 1.4993 | 0.71 | 10 | 1.4576 |
| 1.3484 | 0.79 | 11 | 1.4155 |
| 1.6875 | 0.86 | 12 | 1.3773 |
| 1.2003 | 0.93 | 13 | 1.3420 |
| 1.4851 | 1.0 | 14 | 1.3123 |
| 1.5295 | 1.07 | 15 | 1.2891 |
| 1.0738 | 1.14 | 16 | 1.2648 |
| 1.2737 | 1.21 | 17 | 1.2451 |
| 1.2225 | 1.29 | 18 | 1.2304 |
| 1.0444 | 1.36 | 19 | 1.2194 |
| 1.0923 | 1.43 | 20 | 1.2137 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.0.0
- Datasets 2.16.1
- Tokenizers 0.15.0