metadata
license: gemma
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: google/gemma-2b
model-index:
- name: gemma-2b-for-python-v2
results: []
gemma-2b-for-python-v2
This model is a fine-tuned version of google/gemma-2b on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0070
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: 0.0001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.3988 | 1.0 | 185 | 1.2089 |
| 0.9849 | 2.0 | 370 | 0.9987 |
| 0.8047 | 3.0 | 555 | 0.9539 |
| 0.6775 | 4.0 | 740 | 0.9276 |
| 0.5567 | 5.0 | 925 | 0.9789 |
| 0.4494 | 6.0 | 1110 | 1.0070 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2