finetune_colpali_v1_2-olje_norsk-4bit

This model is a fine-tuned version of vidore/colpaligemma-3b-pt-448-base on the ynuwara/norsk-olje-gass-QnA-ColPali dataset.

Model description

ColPaliGemma-3B model is fine-tuned on a specific question and answering dataset from a book about Oil and Gas in Norwegian language. Fine tuning is done on Colab Pro's NVIDIA A100 GPU.

Intended uses & limitations

This model is suited for RAG uses in oil and gas domain, specifically if the query is in Norwegian

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use 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: 100
• num_epochs: 1.5

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0833	1	0.0837

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.20.3