pali_291311

This model is a fine-tuned version of google/paligemma-3b-pt-224 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8911

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: 2e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
7.2412	0.0425	50	0.9580
6.933	0.0849	100	0.9482
6.7303	0.1274	150	0.9393
6.6387	0.1699	200	0.9324
6.4075	0.2123	250	0.9272
6.8845	0.2548	300	0.9177
6.3864	0.2973	350	0.9143
6.4283	0.3397	400	0.9066
6.8153	0.3822	450	0.9025
6.8832	0.4247	500	0.8956
6.8344	0.4671	550	0.8911

Framework versions

• PEFT 0.14.0
• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1