README.md

metadata

library_name: transformers
tags:
  - mergekit
  - merge
base_model:
  - bunnycore/Llama-3.2-3B-KodCode-R1
  - bunnycore/Llama-3.2-3b-RP-Toxic-R1
  - unsloth/Llama-3.2-3B-Instruct
model-index:
  - name: Llama-3.2-3B-ToxicKod
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: HuggingFaceH4/ifeval
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 63.19
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Llama-3.2-3B-ToxicKod
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: BBH
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 22.98
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Llama-3.2-3B-ToxicKod
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: hendrycks/competition_math
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 16.99
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Llama-3.2-3B-ToxicKod
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 2.13
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Llama-3.2-3B-ToxicKod
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MuSR (0-shot)
          type: TAUR-Lab/MuSR
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 1.43
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Llama-3.2-3B-ToxicKod
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU-PRO (5-shot)
          type: TIGER-Lab/MMLU-Pro
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 20.89
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Llama-3.2-3B-ToxicKod
          name: Open LLM Leaderboard

This model is intended to be uncensored, and it can also role-play and use reasoning.

Reasoning Mode:

Think about the reasoning process in the mind first, then provide the answer.

The reasoning process should be wrapped within <think> </think> tags, then provide the answer after that, i.e., <think> reasoning process here </think> answer.

Models Merged

The following models were included in the merge:

• bunnycore/Llama-3.2-3B-KodCode-R1
• bunnycore/Llama-3.2-3b-RP-Toxic-R1

Configuration

The following YAML configuration was used to produce this model:

models:
  # Pivot model
  - model: unsloth/Llama-3.2-3B-Instruct
  # Target models
  - model: bunnycore/Llama-3.2-3B-KodCode-R1
  - model: bunnycore/Llama-3.2-3b-RP-Toxic-R1
merge_method: sce
base_model: unsloth/Llama-3.2-3B-Instruct
parameters:
  select_topk: 1.2
dtype: bfloat16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	21.27
IFEval (0-Shot)	63.19
BBH (3-Shot)	22.98
MATH Lvl 5 (4-Shot)	16.99
GPQA (0-shot)	2.13
MuSR (0-shot)	1.43
MMLU-PRO (5-shot)	20.89