Havoc999/gemma-4-E2b-it-megacode

Uploaded model

• Developed by: Havoc999
• License: apache-2.0
• Finetuned from model : unsloth/gemma-4-e2b-it-unsloth-bnb-4bit

This gemma4 model was trained 2x faster with Unsloth

This is an adapter only model, you need the base model to work with if you want to use this.

This model was trained on "rombodawg/LosslessMegaCodeTrainingV3_1.6m_Evol" for 100 steps. The model has a loss point of 0.1478. Max Sequence Length: 2048. Training Method: QLoRA (Unsloth FastLanguageModel) targeted across all 7 linear layers (q, k, v, o, gate, up, down).

UPDATE : The merged model : Havoc999/gemma-4-E2b-it-megacode-merged