gemma3-4b-ocrdemo2-fullpage

This model is a fine-tuned version of google/gemma-3-4b-it on the ocrdemo2_train dataset. It achieves the following results on the evaluation set:

• Loss: 0.2236

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: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.3105	1.0	138	0.2719
0.2267	2.0	276	0.2295
0.1875	3.0	414	0.2236

Framework versions

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