bunnycore/Llama-3.2-3B-ToxicKod-0.2

I expect this model to be uncensored and also capable of role-playing, and a little bit of reasoning.

Thinking Mode:

Always think and do evidence-based reasoning, and always reason before any answer. Always self-critique and self-question, and always wrap your thinking in <think><think> answer here

Models Merged

The following models were included in the merge:

• bunnycore/Llama-3.2-3b-RP-Toxic-Fuse
• bunnycore/Llama-3.2-3B-ToxicKod
• bunnycore/Llama-3.2-3b-RP-Toxic-R1
• bunnycore/Llama-3.2-3b-RP-Toxic-Fuse + bunnycore/Llama-3.2-3b-RP-Toxic-R1-lora
• bunnycore/Llama-3.2-3B-KodCode-R1

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: bunnycore/Llama-3.2-3b-RP-Toxic-Fuse+bunnycore/Llama-3.2-3b-RP-Toxic-R1-lora
    parameters:
      weight: 0.3
  - model: bunnycore/Llama-3.2-3b-RP-Toxic-R1
  - model: bunnycore/Llama-3.2-3B-ToxicKod
  - model: bunnycore/Llama-3.2-3B-KodCode-R1
  - model: bunnycore/Llama-3.2-3b-RP-Toxic-Fuse
base_model: unsloth/Llama-3.2-3B-Instruct+bunnycore/Llama-3.2-3b-RP-Toxic-R1-lora
merge_method: model_stock
parameters:
dtype: bfloat16
tokenizer_source: unsloth/Llama-3.2-3B-Instruct