|
|
--- |
|
|
license: apache-2.0 |
|
|
library_name: peft |
|
|
tags: |
|
|
- trl |
|
|
- sft |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- generator |
|
|
base_model: mistralai/Mistral-7B-v0.1 |
|
|
model-index: |
|
|
- name: mistral7b_instruct_generation |
|
|
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. --> |
|
|
|
|
|
# mistral7b_instruct_generation |
|
|
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 1.8070 |
|
|
|
|
|
## 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: 0.03 |
|
|
- training_steps: 500 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:----:|:---------------:| |
|
|
| 1.958 | 0.0 | 20 | 1.8399 | |
|
|
| 1.8756 | 0.01 | 40 | 1.8082 | |
|
|
| 1.888 | 0.01 | 60 | 1.7994 | |
|
|
| 1.7863 | 0.01 | 80 | 1.7892 | |
|
|
| 1.7987 | 0.01 | 100 | 1.7897 | |
|
|
| 1.847 | 0.02 | 120 | 1.7881 | |
|
|
| 1.8505 | 0.02 | 140 | 1.7858 | |
|
|
| 1.7915 | 0.02 | 160 | 1.7900 | |
|
|
| 1.8611 | 0.03 | 180 | 1.7942 | |
|
|
| 1.8827 | 0.03 | 200 | 1.7943 | |
|
|
| 1.88 | 0.03 | 220 | 1.7953 | |
|
|
| 1.8887 | 0.03 | 240 | 1.7888 | |
|
|
| 1.6737 | 0.04 | 260 | 1.7939 | |
|
|
| 1.8471 | 0.04 | 280 | 1.7880 | |
|
|
| 1.8163 | 0.04 | 300 | 1.7835 | |
|
|
| 1.7763 | 0.04 | 320 | 1.7825 | |
|
|
| 1.8567 | 0.05 | 340 | 1.7962 | |
|
|
| 1.8203 | 0.05 | 360 | 1.7871 | |
|
|
| 1.965 | 0.05 | 380 | 1.7897 | |
|
|
| 1.9522 | 0.06 | 400 | 1.8036 | |
|
|
| 1.8826 | 0.06 | 420 | 1.8028 | |
|
|
| 1.8173 | 0.06 | 440 | 1.8205 | |
|
|
| 1.8269 | 0.06 | 460 | 1.8059 | |
|
|
| 1.8498 | 0.07 | 480 | 1.7966 | |
|
|
| 1.9159 | 0.07 | 500 | 1.8070 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.7.1 |
|
|
- Transformers 4.37.2 |
|
|
- Pytorch 2.1.0+cu121 |
|
|
- Datasets 2.16.1 |
|
|
- Tokenizers 0.15.1 |
