Refusal-reduced coder derivative
According to the model card, the model removes refusal behaviors from Qwen/Qwen2.5-Coder-14B-Instruct through directional intervention.
Open Source Model Profile · BlossomsAI
Qwen2.5-Coder-14B-Instruct-Uncensored
Qwen2.5-Coder-14B-Instruct-Uncensored is a 14.77B-parameter Qwen2 coder model from BlossomsAI focused on removing refusal behaviors.
- Publisher
- BlossomsAI
- Task
- text-generation
- Model type
- qwen2
- License
- mit
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
Qwen2.5-Coder-14B-Instruct-Uncensored is published by BlossomsAI as a Qwen2 text-generation model. The captured configuration identifies Qwen2ForCausalLM and Safetensors metadata reports 14,770,033,664 parameters. According to the model card, it modifies Qwen/Qwen2.5-Coder-14B-Instruct to reduce refusals, with card data recording mit.
Recorded capabilities
Publisher-cited refusal research
According to the model card, implementation follows research findings from Refusal in LLMs is Mediated by a Single Direction.
GGFU quantization note
According to the model card, pre-quantized models are available in GGFU format for resource-efficient deployment.
Use cases in the source record
- Code-oriented conversational generation using the publisher's documented refusal-reduced Qwen2.5-Coder setup.
- Resource-constrained reuse where the publisher's documented GGFU quantized versions are relevant.
Limitations and unknowns
- No evaluation results, context-window value, or VRAM requirement was extracted.
- Refusal-removal behavior and quantization details come only from the publisher's brief description and were not independently verified.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
Source and provenance
Source: BlossomsAI/Qwen2.5-Coder-14B-Instruct-Uncensored
Captured: Unknown. Processed: 2026-09-07T19:35:03.108784+00:00.
🎯 Qwen2.5-Coder-14B-Instruct-Uncensored An optimized version of Qwen2.5-Coder-14B-Instruct focused on removing refusal behaviors 📌 Overview A streamlined implementation that removes refusal behaviors from Qwen/Qwen2.5-Coder-14B-Instruct , based on directional intervention techniques. 🔍 Key Features Implements directional intervention for behavior modification Fully quantized for resource-efficient deployment 🛠️ Implementation Based on research findings from "Refusal in LLMs is Mediated by a Single Direction" 💫 Quantization Pre-quantized models available in GGFU format: 🤝 Contributors Developed with ❤️ by BlossomAI Star ⭐️ this…
F001F002F003F004F005F006F007F009F010F011F012F015