README.md

metadata

license: other
tags:
  - yi
  - moe
license_name: yi-license
license_link: https://huggingface.co./01-ai/Yi-34B-200K/blob/main/LICENSE
model-index:
  - name: Astralis-4x34B
    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: 69.71
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Astralis-4x34B
          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: 85.17
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Astralis-4x34B
          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: 77.24
            name: accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Astralis-4x34B
          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: 63.55
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Astralis-4x34B
          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: 84.14
            name: accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Astralis-4x34B
          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: 71.65
            name: accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Astralis-4x34B
          name: Open LLM Leaderboard

More details coming soon.

Yaml Config to reproduce

base_model: nontoxic-bagel-34b-v0.2
gate_mode: hidden
dtype: bfloat16

experts:
  - source_model: bagel-dpo-34b-v0.2
    positive_prompts: ["question answering", "Q:", science", "biology", "chemistry", "physics"]
    negative_prompts: ["math", "reason", "mathematics", "solve", "count", "code", "python", "javascript", "programming", "algorithm"]

  - source_model: Nous-Hermes-2-Yi-34B
    positive_prompts: ["chat", "math", "reason", "mathematics", "solve", "count", "python", "javascript", "programming", "algorithm", "tell me", "assistant"]

  - source_model: SUS-Chat-34B
    positive_prompts: ["math", "reason", "mathematics", "solve", "count", "assistant"]

  - source_model: yi-34B-v3
    positive_prompts: [""]

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	75.24
AI2 Reasoning Challenge (25-Shot)	69.71
HellaSwag (10-Shot)	85.17
MMLU (5-Shot)	77.24
TruthfulQA (0-shot)	63.55
Winogrande (5-shot)	84.14
GSM8k (5-shot)	71.65