README.md

metadata

license:
  - other
tags:
  - dare ties
model-index:
  - name: DARE_TIES_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: 74.32
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=yunconglong/DARE_TIES_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: 89.5
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=yunconglong/DARE_TIES_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: 64.47
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=yunconglong/DARE_TIES_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: 78.66
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=yunconglong/DARE_TIES_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: 88.08
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=yunconglong/DARE_TIES_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: 67.55
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=yunconglong/DARE_TIES_13B
          name: Open LLM Leaderboard

DARE_TIES_13B

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B as a base.

Models Merged

The following models were included in the merge:

• ./13B_DPO
• ./13B_MATH_DPO

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B
    parameters:
      density: 1.0
      weight: 1.0
  - model: ./13B_MATH_DPO
    parameters:
      density: 0.5
      weight: [0.33, 0.4, 0.33]
  - model: ./13B_DPO
    parameters:
      density: [0.33, 0.45, 0.66]
      weight: 0.66
merge_method: dare_ties
base_model: yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16
tokenizer_source : union

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	77.10
AI2 Reasoning Challenge (25-Shot)	74.32
HellaSwag (10-Shot)	89.50
MMLU (5-Shot)	64.47
TruthfulQA (0-shot)	78.66
Winogrande (5-shot)	88.08
GSM8k (5-shot)	67.55