|
--- |
|
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-v4 |
|
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-v4 |
|
|
|
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: 0.4623 |
|
|
|
## 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 | 0.7933 | |
|
| 0.8309 | 0.17 | 10 | 0.7250 | |
|
| 0.6972 | 0.26 | 15 | 0.6792 | |
|
| 0.6841 | 0.34 | 20 | 0.6448 | |
|
| 0.645 | 0.43 | 25 | 0.6158 | |
|
| 0.626 | 0.52 | 30 | 0.5929 | |
|
| 0.5645 | 0.6 | 35 | 0.5719 | |
|
| 0.5722 | 0.69 | 40 | 0.5545 | |
|
| 0.5489 | 0.78 | 45 | 0.5385 | |
|
| 0.5206 | 0.86 | 50 | 0.5283 | |
|
| 0.4599 | 0.95 | 55 | 0.5171 | |
|
| 0.5232 | 1.03 | 60 | 0.5082 | |
|
| 0.4798 | 1.12 | 65 | 0.5032 | |
|
| 0.3585 | 1.21 | 70 | 0.4984 | |
|
| 0.3923 | 1.29 | 75 | 0.4899 | |
|
| 0.3915 | 1.38 | 80 | 0.4825 | |
|
| 0.3845 | 1.47 | 85 | 0.4758 | |
|
| 0.3768 | 1.55 | 90 | 0.4752 | |
|
| 0.3928 | 1.64 | 95 | 0.4668 | |
|
| 0.3986 | 1.72 | 100 | 0.4632 | |
|
| 0.3495 | 1.81 | 105 | 0.4607 | |
|
| 0.4014 | 1.9 | 110 | 0.4563 | |
|
| 0.3902 | 1.98 | 115 | 0.4519 | |
|
| 0.3081 | 2.07 | 120 | 0.4656 | |
|
| 0.3204 | 2.16 | 125 | 0.4569 | |
|
| 0.2844 | 2.24 | 130 | 0.4605 | |
|
| 0.2501 | 2.33 | 135 | 0.4595 | |
|
| 0.2723 | 2.41 | 140 | 0.4547 | |
|
| 0.2979 | 2.5 | 145 | 0.4662 | |
|
| 0.2884 | 2.59 | 150 | 0.4548 | |
|
| 0.2944 | 2.67 | 155 | 0.4587 | |
|
| 0.2575 | 2.76 | 160 | 0.4542 | |
|
| 0.2558 | 2.84 | 165 | 0.4499 | |
|
| 0.2165 | 2.93 | 170 | 0.4511 | |
|
| 0.2806 | 3.02 | 175 | 0.4484 | |
|
| 0.1799 | 3.1 | 180 | 0.4799 | |
|
| 0.1877 | 3.19 | 185 | 0.4608 | |
|
| 0.1918 | 3.28 | 190 | 0.4738 | |
|
| 0.1812 | 3.36 | 195 | 0.4665 | |
|
| 0.199 | 3.45 | 200 | 0.4714 | |
|
| 0.1581 | 3.53 | 205 | 0.4699 | |
|
| 0.1918 | 3.62 | 210 | 0.4613 | |
|
| 0.2052 | 3.71 | 215 | 0.4667 | |
|
| 0.1893 | 3.79 | 220 | 0.4626 | |
|
| 0.2177 | 3.88 | 225 | 0.4606 | |
|
| 0.2196 | 3.97 | 230 | 0.4623 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
