|
--- |
|
license: apache-2.0 |
|
base_model: mistralai/Mistral-7B-Instruct-v0.2 |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
model-index: |
|
- name: Summary_M2_1000steps_1e7rate_SFT2 |
|
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. --> |
|
|
|
# Summary_M2_1000steps_1e7rate_SFT2 |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5731 |
|
|
|
## 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: 1e-07 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- training_steps: 1000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.8232 | 0.2 | 50 | 1.7949 | |
|
| 1.53 | 0.4 | 100 | 1.4490 | |
|
| 1.0256 | 0.6 | 150 | 0.9395 | |
|
| 0.6284 | 0.8 | 200 | 0.6030 | |
|
| 0.5923 | 1.0 | 250 | 0.5887 | |
|
| 0.5851 | 1.2 | 300 | 0.5835 | |
|
| 0.675 | 1.4 | 350 | 0.5802 | |
|
| 0.5931 | 1.6 | 400 | 0.5780 | |
|
| 0.5629 | 1.8 | 450 | 0.5765 | |
|
| 0.6207 | 2.0 | 500 | 0.5753 | |
|
| 0.5956 | 2.2 | 550 | 0.5745 | |
|
| 0.5902 | 2.4 | 600 | 0.5740 | |
|
| 0.6132 | 2.6 | 650 | 0.5736 | |
|
| 0.5964 | 2.8 | 700 | 0.5733 | |
|
| 0.5844 | 3.0 | 750 | 0.5732 | |
|
| 0.6054 | 3.2 | 800 | 0.5731 | |
|
| 0.6079 | 3.4 | 850 | 0.5731 | |
|
| 0.6549 | 3.6 | 900 | 0.5731 | |
|
| 0.5838 | 3.8 | 950 | 0.5731 | |
|
| 0.5724 | 4.0 | 1000 | 0.5731 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.41.2 |
|
- Pytorch 2.0.0+cu117 |
|
- Datasets 2.19.2 |
|
- Tokenizers 0.19.1 |
|
|