README.md

metadata

language:
  - en
license: agpl-3.0
model-index:
  - name: Michel-13B
    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: 61.26
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/Michel-13B
          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: 83.21
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/Michel-13B
          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: 55.05
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/Michel-13B
          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: 50.43
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/Michel-13B
          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: 75.22
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/Michel-13B
          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: 20.17
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/Michel-13B
          name: Open LLM Leaderboard

Michel - 13B

A Uncensored fine tune model general tasks focused of NousHermes-Llama2-13B.
With the help of my team.

Quants

Michel Exl2 Quant
Michel GGUF Quant

Useful

• Uses Llama2 prompt template.
• Next one will be more interesting :}

I had good results with this parameters:

• temperature: 0.8
• top_p: 0.75
• min_p: 0
• top_k: 0
• repetition_penalty: 1.05

General Usecase Test

BenchMarks on OpenLLM Leaderboard

More details: Michel-13B OpenLLM BenchMarks

Thank you h2m for the compute

Have Fun :)

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	57.56
AI2 Reasoning Challenge (25-Shot)	61.26
HellaSwag (10-Shot)	83.21
MMLU (5-Shot)	55.05
TruthfulQA (0-shot)	50.43
Winogrande (5-shot)	75.22
GSM8k (5-shot)	20.17