DavidAU/gemma-3-12b-it-vl-Polaris-Heretic-Uncensored-Thinking
The DavidAU/gemma-3-12b-it-vl-Polaris-Heretic-Uncensored-Thinking model is a 12 billion parameter Gemma-based instruction-tuned language model, fine-tuned for uncensored, deep reasoning capabilities. It features a 128k context window and is specifically optimized to provide direct, detailed responses without refusal, enhancing both general operation and image processing. This model excels in scenarios requiring explicit and unconstrained content generation, offering stable reasoning across a wide temperature range.
Loading preview...
Model Overview
DavidAU/gemma-3-12b-it-vl-Polaris-Heretic-Uncensored-Thinking is a 12 billion parameter Gemma-based instruction-tuned model, distinguished by its fully uncensored and deep reasoning capabilities. Fine-tuned using the Polaris reasoning dataset via Unsloth, this model aims to provide direct and detailed responses without the typical refusals found in other models. It boasts a 128k context window and maintains reasoning stability across a wide temperature range (0.1 to 2.5).
Key Capabilities & Differentiators
- Uncensored Output: Designed to generate content exactly as requested, including explicit or graphic material, by requiring specific directives (e.g., using slang or explicit terms).
- Enhanced Reasoning: Integrates deep thinking into general model operation, output generation, and image processing, leading to compact yet highly detailed reasoning.
- Flexible Activation: Reasoning is often automatic but can be explicitly activated using "think deeply: prompt" or through specialized Jinja templates and system prompts.
- Performance: Benchmarks show improved performance over its Heretic, uncensored base, particularly in
arc_challenge,hellaswag, andwinograndetasks. It also demonstrates a significantly reduced refusal rate (7/100) compared to the original Google Gemma model (98/100) with a low KL divergence of 0.0826, indicating minimal damage to the model during de-censoring.
Use Cases
This model is ideal for applications requiring:
- Unrestricted Content Generation: For scenarios where explicit, graphic, or uncensored text is required, provided appropriate directives are given.
- Deep and Detailed Reasoning: When complex problem-solving or highly analytical responses are needed, benefiting from its enhanced reasoning capabilities.
- Creative Writing & Roleplay: Especially for generating content that might be filtered by other models, allowing for greater creative freedom.
Optimal performance can be achieved by adjusting settings like Smoothing_factor (to 1.5) and Repetition Penalty (1.1-1.15) in compatible interfaces.