license, language, base_model, tags, pipeline_tag, library_name, inference

license	language	base_model	tags	pipeline_tag	library_name	inference
apache-2.0	en	Qwen/Qwen2.5-Coder-7B-Instruct	text-generation abliterated bruno heretic decensored optuna-optimized qwen2 code conversational uncensored	text-generation	transformers	parameters max_new_tokens temperature top_p 512 0.7 0.9

Qwen2.5-Coder-7B-Instruct-Bruno (Abliterated)

This is an abliterated version of Qwen/Qwen2.5-Coder-7B-Instruct with reduced refusals.

Model Details

• Base Model: Qwen/Qwen2.5-Coder-7B-Instruct

• Modification: Abliteration (refusal direction removal)

• Architecture: Qwen2ForCausalLM

• Parameters: 7B

• Context Length: 32,768 tokens

Usage

License

Apache 2.0 (same as base model)