|
--- |
|
license: gemma |
|
library_name: peft |
|
tags: |
|
- generated_from_trainer |
|
base_model: google/gemma-2b-it |
|
model-index: |
|
- name: gemma-2b-it-retrained_network_corpus |
|
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. --> |
|
|
|
# gemma-2b-it-retrained_network_corpus |
|
|
|
This model is a fine-tuned version of [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.3035 |
|
|
|
## 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: 1 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 2 |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 2.9983 | 1.0 | 28 | 2.3757 | |
|
| 2.1934 | 2.0 | 56 | 2.3035 | |
|
| 1.6321 | 3.0 | 84 | 2.4507 | |
|
| 1.1753 | 4.0 | 112 | 2.6984 | |
|
| 0.8011 | 5.0 | 140 | 2.9980 | |
|
| 0.5162 | 6.0 | 168 | 3.2588 | |
|
| 0.304 | 7.0 | 196 | 3.7305 | |
|
| 0.1822 | 8.0 | 224 | 4.0858 | |
|
| 0.1148 | 9.0 | 252 | 4.3105 | |
|
| 0.0855 | 10.0 | 280 | 4.5450 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.11.1 |
|
- Transformers 4.41.2 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.19.2 |
|
- Tokenizers 0.19.1 |
