README.md

metadata

language:
  - en
license: llama3.2
base_model:
  - meta-llama/Llama-3.2-3B-Instruct
model-index:
  - name: LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
    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: 69.31
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
          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: 23.81
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
          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: 10.42
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
          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: 3.24
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
          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: 4.05
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
          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: 23.64
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=PJMixers-Dev/LLaMa-3.2-Instruct-JankMix-v0.1-SFT-3B
          name: Open LLM Leaderboard

Quick test tune overtop of meta-llama/Llama-3.2-3B-Instruct using a ~50/50 mix of instruct and completion data.

Note: Training nowhere near complete so I'm unsure how strong of an effect it had. Still refuses requests like meta-llama/Llama-3.2-3B-Instruct.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	22.41
IFEval (0-Shot)	69.31
BBH (3-Shot)	23.81
MATH Lvl 5 (4-Shot)	10.42
GPQA (0-shot)	3.24
MuSR (0-shot)	4.05
MMLU-PRO (5-shot)	23.64