mahdiboughrous/safe-flat-lora-baseline-gemma3-1b-it-merged
Part of the Safe-Flat-LoRA project (the Safety-Flatness Paradox in 4-bit
quantization of LoRA-fine-tuned LLMs -- see the project blueprint for the
full motivation and methodology).
- Base model:
google/gemma-3-1b-it
- Method:
baseline (plain LoRA fine-tune, used as the pre-Flat-LoRA baseline)
- Artifact: merged
- Fine-tuning task: SQL generation (
b-mc2/sql-create-context), used as a
downstream-task proxy to study how post-training 4-bit (NF4) quantization
shifts safety-refusal behavior after LoRA fine-tuning.
Base model Gemma-3-1B-it is under Google's Gemma Terms of Use. Redistributing derivatives requires complying with the Gemma Prohibited Use Policy -- this repo inherits those terms, it is not independently licensed.
