dushj98/action-gemma-2-test-v2

ActionGemma2 v2

• Developed by: dushj98
• License: apache-2.0
• Finetuned from model : unsloth/gemma-2-2b-it-bnb-4bit

This gemma2 model was trained 2x faster with Unsloth and Huggingface's TRL library.

Note

This version of the model was fine-tuned for a full epoch whereas ActionGemma2 Test v1 was only fine-tuned for 60 steps.

This is a fine-tuned model for testing purposes, fine-tuned for a full single epoch in Kaggle. The dataset used for SFT has been changed since and this model is no longer valid.