---
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

```yaml
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](https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co./datasets/open-llm-leaderboard/details_Weyaxi__Astralis-4x34B)

|             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|