gemma3-4b-journal-field

This model is a fine-tuned version of google/gemma-3-4b-it on the journal_train dataset. It achieves the following results on the evaluation set:

• Loss: 0.0238

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
0.1761	0.6504	20	0.0735
0.0638	1.2927	40	0.0420
0.0587	1.9431	60	0.0508
0.0316	2.5854	80	0.0321
0.0142	3.2276	100	0.0276
0.0083	3.8780	120	0.0249
0.005	4.5203	140	0.0238

Framework versions

• PEFT 0.15.2
• Transformers 4.52.4
• Pytorch 2.10.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.1