README.md

metadata

license: unknown
model-index:
  - name: contaminated_proof_7b_v1.0_safetensor
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: AI2 Reasoning Challenge (25-Shot)
          type: ai2_arc
          config: ARC-Challenge
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: acc_norm
            value: 78.07
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Contamination/contaminated_proof_7b_v1.0_safetensor
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HellaSwag (10-Shot)
          type: hellaswag
          split: validation
          args:
            num_few_shot: 10
        metrics:
          - type: acc_norm
            value: 90.22
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Contamination/contaminated_proof_7b_v1.0_safetensor
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU (5-Shot)
          type: cais/mmlu
          config: all
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 78.92
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Contamination/contaminated_proof_7b_v1.0_safetensor
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: TruthfulQA (0-shot)
          type: truthful_qa
          config: multiple_choice
          split: validation
          args:
            num_few_shot: 0
        metrics:
          - type: mc2
            value: 82.29
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Contamination/contaminated_proof_7b_v1.0_safetensor
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Winogrande (5-shot)
          type: winogrande
          config: winogrande_xl
          split: validation
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 88.16
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Contamination/contaminated_proof_7b_v1.0_safetensor
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GSM8k (5-shot)
          type: gsm8k
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 69.14
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Contamination/contaminated_proof_7b_v1.0_safetensor
          name: Open LLM Leaderboard

This model has same weights with Contamination/contaminated_proof_7b_v1.0

WARNING: Contamination

This model is TOTALLY CONTAMINATED, which made resulting model unreliable.

SO DO NOT USE THIS MODEL FOR ANY PURPOSE. PLEASE ONLY USE FOR REFERENCE.

This model is trained with ultrachat_200k data to have conversational features.

MODEL ARCHITECTURE

This model was initialized with Mistral-7B-v0.1

PLEASE NOTE

Users and sponsors should be wary that many models are also unreliable. I hope our model can show the vulnerablity of the leaderboard.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	81.14
AI2 Reasoning Challenge (25-Shot)	78.07
HellaSwag (10-Shot)	90.22
MMLU (5-Shot)	78.92
TruthfulQA (0-shot)	82.29
Winogrande (5-shot)	88.16
GSM8k (5-shot)	69.14