|
--- |
|
base_model: Unbabel/TowerInstruct-7B-v0.2 |
|
library_name: peft |
|
license: cc-by-nc-4.0 |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
model-index: |
|
- name: BioTowerInstruct-7b-v0.2-run09 |
|
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. --> |
|
|
|
# BioTowerInstruct-7b-v0.2-run09 |
|
|
|
This model is a fine-tuned version of [Unbabel/TowerInstruct-7B-v0.2](https://huggingface.co./Unbabel/TowerInstruct-7B-v0.2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.3265 |
|
|
|
## 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-05 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 2.1703 | 0.08 | 30 | 2.0597 | |
|
| 1.7574 | 0.16 | 60 | 1.8526 | |
|
| 1.5809 | 0.24 | 90 | 1.7084 | |
|
| 1.3437 | 0.32 | 120 | 1.5682 | |
|
| 1.3627 | 0.4 | 150 | 1.5074 | |
|
| 1.2534 | 0.48 | 180 | 1.4496 | |
|
| 1.1732 | 0.56 | 210 | 1.4031 | |
|
| 1.1735 | 0.64 | 240 | 1.3727 | |
|
| 1.1581 | 0.72 | 270 | 1.3522 | |
|
| 1.2024 | 0.8 | 300 | 1.3389 | |
|
| 1.1346 | 0.88 | 330 | 1.3315 | |
|
| 1.0714 | 0.96 | 360 | 1.3265 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.0 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
