|
--- |
|
license: apache-2.0 |
|
base_model: mistralai/Mistral-7B-Instruct-v0.1 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: radia-fine-tune-mistral-7b-lora |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# radia-fine-tune-mistral-7b-lora |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4616 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 16 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: constant |
|
- lr_scheduler_warmup_ratio: 0.3 |
|
- num_epochs: 4 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.038 | 0.09 | 5 | 0.7930 | |
|
| 0.8312 | 0.17 | 10 | 0.7245 | |
|
| 0.6972 | 0.26 | 15 | 0.6791 | |
|
| 0.6848 | 0.34 | 20 | 0.6456 | |
|
| 0.6457 | 0.43 | 25 | 0.6159 | |
|
| 0.6261 | 0.52 | 30 | 0.5927 | |
|
| 0.5639 | 0.6 | 35 | 0.5718 | |
|
| 0.5723 | 0.69 | 40 | 0.5540 | |
|
| 0.5473 | 0.78 | 45 | 0.5389 | |
|
| 0.5209 | 0.86 | 50 | 0.5284 | |
|
| 0.4591 | 0.95 | 55 | 0.5177 | |
|
| 0.5233 | 1.03 | 60 | 0.5080 | |
|
| 0.4805 | 1.12 | 65 | 0.5030 | |
|
| 0.3604 | 1.21 | 70 | 0.4987 | |
|
| 0.3927 | 1.29 | 75 | 0.4907 | |
|
| 0.3934 | 1.38 | 80 | 0.4849 | |
|
| 0.3859 | 1.47 | 85 | 0.4761 | |
|
| 0.3779 | 1.55 | 90 | 0.4748 | |
|
| 0.3929 | 1.64 | 95 | 0.4655 | |
|
| 0.3973 | 1.72 | 100 | 0.4635 | |
|
| 0.3512 | 1.81 | 105 | 0.4599 | |
|
| 0.4027 | 1.9 | 110 | 0.4575 | |
|
| 0.3917 | 1.98 | 115 | 0.4504 | |
|
| 0.3069 | 2.07 | 120 | 0.4667 | |
|
| 0.3203 | 2.16 | 125 | 0.4569 | |
|
| 0.2807 | 2.24 | 130 | 0.4615 | |
|
| 0.2471 | 2.33 | 135 | 0.4612 | |
|
| 0.2724 | 2.41 | 140 | 0.4553 | |
|
| 0.2976 | 2.5 | 145 | 0.4665 | |
|
| 0.2873 | 2.59 | 150 | 0.4551 | |
|
| 0.2968 | 2.67 | 155 | 0.4568 | |
|
| 0.2577 | 2.76 | 160 | 0.4564 | |
|
| 0.2569 | 2.84 | 165 | 0.4496 | |
|
| 0.2167 | 2.93 | 170 | 0.4486 | |
|
| 0.2785 | 3.02 | 175 | 0.4518 | |
|
| 0.176 | 3.1 | 180 | 0.4798 | |
|
| 0.1909 | 3.19 | 185 | 0.4588 | |
|
| 0.1883 | 3.28 | 190 | 0.4768 | |
|
| 0.1806 | 3.36 | 195 | 0.4693 | |
|
| 0.1998 | 3.45 | 200 | 0.4732 | |
|
| 0.1573 | 3.53 | 205 | 0.4745 | |
|
| 0.1908 | 3.62 | 210 | 0.4627 | |
|
| 0.2042 | 3.71 | 215 | 0.4695 | |
|
| 0.1918 | 3.79 | 220 | 0.4620 | |
|
| 0.2163 | 3.88 | 225 | 0.4574 | |
|
| 0.2189 | 3.97 | 230 | 0.4616 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
