|
--- |
|
license: apache-2.0 |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
base_model: mistralai/Mistral-7B-Instruct-v0.2 |
|
model-index: |
|
- name: mistral7binstruct_summarize |
|
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. --> |
|
|
|
# mistral7binstruct_summarize |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.4924 |
|
|
|
## 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: 0.0005 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 3 |
|
- training_steps: 300 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.6188 | 0.22 | 25 | 1.5050 | |
|
| 1.5803 | 0.43 | 50 | 1.4716 | |
|
| 1.541 | 0.65 | 75 | 1.4659 | |
|
| 1.5501 | 0.86 | 100 | 1.4611 | |
|
| 1.3937 | 1.08 | 125 | 1.4710 | |
|
| 1.388 | 1.29 | 150 | 1.4705 | |
|
| 1.3811 | 1.51 | 175 | 1.4681 | |
|
| 1.2997 | 1.72 | 200 | 1.4695 | |
|
| 1.3746 | 1.94 | 225 | 1.4583 | |
|
| 1.2607 | 2.16 | 250 | 1.4895 | |
|
| 1.172 | 2.37 | 275 | 1.4936 | |
|
| 1.1847 | 2.59 | 300 | 1.4924 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.9.0 |
|
- Transformers 4.38.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
