|
--- |
|
license: apache-2.0 |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
base_model: teknium/OpenHermes-2.5-Mistral-7B |
|
model-index: |
|
- name: paper_imp_new_format_1 |
|
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. --> |
|
|
|
# paper_imp_new_format_1 |
|
|
|
This model is a fine-tuned version of [teknium/OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.6991 |
|
|
|
## 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: 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: 200 |
|
- num_epochs: 4 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.8356 | 0.33 | 9 | 1.8214 | |
|
| 1.8011 | 0.67 | 18 | 1.7745 | |
|
| 1.724 | 1.0 | 27 | 1.6848 | |
|
| 1.6316 | 1.33 | 36 | 1.5849 | |
|
| 1.5605 | 1.67 | 45 | 1.4827 | |
|
| 1.4149 | 2.0 | 54 | 1.3548 | |
|
| 1.2969 | 2.33 | 63 | 1.1693 | |
|
| 0.9925 | 2.67 | 72 | 0.9323 | |
|
| 0.8631 | 3.0 | 81 | 0.8125 | |
|
| 0.7053 | 3.33 | 90 | 0.7490 | |
|
| 0.6931 | 3.67 | 99 | 0.7222 | |
|
| 0.6993 | 4.0 | 108 | 0.6991 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.7.1 |
|
- Transformers 4.36.2 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.15.0 |
