metadata
license: gemma
base_model: google/gemma-2-2b
tags:
- easylm
- trl
- sft
- generated_from_trainer
- trl
- sft
- generated_from_trainer
datasets:
- alpaca_farm
model-index:
- name: easylm-sft-gemma-2-2b
results: []
easylm-sft-gemma-2-2b
This model is a fine-tuned version of google/gemma-2-2b on the alpaca_farm dataset. It achieves the following results on the evaluation set:
- Loss: 0.7072
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: 3e-06
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 16
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6727 | 0.16 | 100 | 0.6869 |
| 0.6056 | 0.32 | 200 | 0.6831 |
| 0.7033 | 0.48 | 300 | 0.6797 |
| 0.6786 | 0.64 | 400 | 0.6771 |
| 0.6476 | 0.8 | 500 | 0.6736 |
| 0.6562 | 0.96 | 600 | 0.6708 |
| 0.461 | 1.12 | 700 | 0.7041 |
| 0.4578 | 1.28 | 800 | 0.7093 |
| 0.4817 | 1.44 | 900 | 0.7055 |
| 0.4324 | 1.6 | 1000 | 0.7080 |
| 0.4693 | 1.76 | 1100 | 0.7081 |
| 0.4475 | 1.92 | 1200 | 0.7070 |
Framework versions
- Transformers 4.43.3
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1