README.md

metadata

license: cc-by-4.0
tags:
  - mistral
  - merge
pipeline_tag: text-generation
model-index:
  - name: xDAN-SlimOrca
    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: 65.61
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Azazelle/xDAN-SlimOrca
          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: 85.7
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Azazelle/xDAN-SlimOrca
          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: 63.67
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Azazelle/xDAN-SlimOrca
          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: 57.68
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Azazelle/xDAN-SlimOrca
          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: 77.66
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Azazelle/xDAN-SlimOrca
          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: 57.92
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Azazelle/xDAN-SlimOrca
          name: Open LLM Leaderboard

Model Card for xDAN-SlimOrca

Slerp merge of xDAN-L1-Chat-RL-v1 and mistral-7b-slimorcaboros.

.yaml file for mergekit

slices:
  - sources:
      - model: xDAN-AI/xDAN-L1-Chat-RL-v1
        layer_range: [0, 32]
      - model: openaccess-ai-collective/mistral-7b-slimorcaboros
        layer_range: [0, 32]
merge_method: slerp
base_model: mistralai/Mistral-7B-v0.1
parameters:
  t:
    - filter: self_attn
      value: [0.14, 0.57, 0.4, 0.74, 1]
    - filter: mlp
      value: [0.86, 0.43, 0.6, 0.26, 0]
    - value: 0.5 # fallback for rest of tensors
dtype: float16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	68.04
AI2 Reasoning Challenge (25-Shot)	65.61
HellaSwag (10-Shot)	85.70
MMLU (5-Shot)	63.67
TruthfulQA (0-shot)	57.68
Winogrande (5-shot)	77.66
GSM8k (5-shot)	57.92