metadata
license: other
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: google/gemma-2b
model-index:
- name: Mental-Gemma-2b
results: []
Mental-Gemma-2b
This model is a fine-tuned version of google/gemma-2b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.5483
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.9567 | 0.79 | 100 | 2.9099 |
| 2.8576 | 1.59 | 200 | 2.8171 |
| 2.7521 | 2.38 | 300 | 2.7228 |
| 2.6615 | 3.17 | 400 | 2.6415 |
| 2.6034 | 3.97 | 500 | 2.5863 |
| 2.555 | 4.76 | 600 | 2.5637 |
| 2.5408 | 5.56 | 700 | 2.5563 |
| 2.5204 | 6.35 | 800 | 2.5526 |
| 2.5319 | 7.14 | 900 | 2.5503 |
| 2.5336 | 7.94 | 1000 | 2.5491 |
| 2.5264 | 8.73 | 1100 | 2.5485 |
| 2.5166 | 9.52 | 1200 | 2.5483 |
Framework versions
- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2