gemma3-4b-field-extraction
This model is a fine-tuned version of google/gemma-3-4b-it on the field_extraction_train dataset. It achieves the following results on the evaluation set:
- Loss: 0.1873
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Use 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 |
|---|---|---|---|
| 0.7081 | 1.0 | 14 | 0.4258 |
| 0.1982 | 2.0 | 28 | 0.2351 |
| 0.0935 | 3.0 | 42 | 0.2017 |
| 0.0526 | 4.0 | 56 | 0.1896 |
| 0.0398 | 5.0 | 70 | 0.1873 |
Framework versions
- PEFT 0.15.2
- Transformers 4.52.4
- Pytorch 2.10.0+cu128
- Datasets 3.6.0
- Tokenizers 0.21.1
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support