metadata
license: other
base_model: google/gemma-2b
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: gemma_2b_scotus
results: []
gemma_2b_scotus
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: 1.6088
- Accuracy: 0.5186
- F1 Macro: 0.3274
- F1 Micro: 0.5186
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: 5e-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
|---|---|---|---|---|---|---|
| 2.2227 | 0.32 | 50 | 2.1669 | 0.31 | 0.0991 | 0.31 |
| 1.7367 | 0.64 | 100 | 1.8375 | 0.425 | 0.2124 | 0.425 |
| 1.6836 | 0.96 | 150 | 1.6646 | 0.4836 | 0.2551 | 0.4836 |
| 1.1611 | 1.27 | 200 | 1.8198 | 0.4386 | 0.2719 | 0.4386 |
| 1.0922 | 1.59 | 250 | 1.7039 | 0.49 | 0.2888 | 0.49 |
| 1.0527 | 1.91 | 300 | 1.6088 | 0.5186 | 0.3274 | 0.5186 |
| 0.5763 | 2.23 | 350 | 1.7765 | 0.4929 | 0.3462 | 0.4929 |
| 0.4645 | 2.55 | 400 | 1.7984 | 0.4986 | 0.3434 | 0.4986 |
| 0.394 | 2.87 | 450 | 1.7742 | 0.4993 | 0.3472 | 0.4993 |
Framework versions
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2