metadata
license: gemma
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: google/gemma-2b
model-index:
- name: gemma-7b-gem
results: []
gemma-7b-gem
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: 1.1198
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: 2.5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 15
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.016 | 2 | 1.1601 |
| No log | 0.032 | 4 | 1.1470 |
| 1.1276 | 0.048 | 6 | 1.1375 |
| 1.1276 | 0.064 | 8 | 1.1304 |
| 1.1805 | 0.08 | 10 | 1.1252 |
| 1.1805 | 0.096 | 12 | 1.1217 |
| 1.1805 | 0.112 | 14 | 1.1198 |
Framework versions
- PEFT 0.10.1.dev0
- Transformers 4.40.0.dev0
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2