metadata
library_name: peft
license: gemma
base_model: google/gemma-7b
tags:
- trl
- sft
- generated_from_trainer
datasets:
- generator
model-index:
- name: gemma7b-lora-alpaca-11-v1
results: []
gemma7b-lora-alpaca-11-v1
This model is a fine-tuned version of google/gemma-7b on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 1.6643
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: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- total_eval_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.9056 | 0.9924 | 65 | 2.6113 |
| 1.8271 | 2.0 | 131 | 1.8230 |
| 1.7019 | 2.9924 | 196 | 1.7041 |
| 1.7024 | 4.0 | 262 | 1.6962 |
| 1.6463 | 4.9618 | 325 | 1.6643 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.2
- Pytorch 2.3.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3