README.md

metadata

license: apache-2.0
tags:
  - merge
  - mergekit
  - lazymergekit
  - dvilasuero/DistilabelBeagle14-7B
  - teknium/OpenHermes-2.5-Mistral-7B
model-index:
  - name: DistilabelCerberus-7B-slerp
    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: 68.17
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Stopwolf/DistilabelCerberus-7B-slerp
          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: 86.78
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Stopwolf/DistilabelCerberus-7B-slerp
          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: 64.2
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Stopwolf/DistilabelCerberus-7B-slerp
          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: 60.93
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Stopwolf/DistilabelCerberus-7B-slerp
          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: 79.48
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Stopwolf/DistilabelCerberus-7B-slerp
          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: 69.83
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Stopwolf/DistilabelCerberus-7B-slerp
          name: Open LLM Leaderboard

DistilabelCerberus-7B-slerp

DistilabelCerberus-7B-slerp is a merge of the following models using mergekit:

• dvilasuero/DistilabelBeagle14-7B
• teknium/OpenHermes-2.5-Mistral-7B

🧩 Configuration

slices:
  - sources:
      - model: dvilasuero/DistilabelBeagle14-7B
        layer_range: [0, 32]
      - model: teknium/OpenHermes-2.5-Mistral-7B
        layer_range: [0, 32]
merge_method: slerp
base_model: teknium/OpenHermes-2.5-Mistral-7B
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16

Results

	ARC-C	Hellaswag	ThruthfulQA	Winogrande	GSM8K
OpenHermes-2.5-Mistral-7B	61.26	65.22	52.24	78.06	26.08
DistilabelBeagle14-7B	?	?	71.66	?	?
DistilabelCerberus-7B-slerp	65.44	69.29	60.93	79.48	69.82

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	71.56
AI2 Reasoning Challenge (25-Shot)	68.17
HellaSwag (10-Shot)	86.78
MMLU (5-Shot)	64.20
TruthfulQA (0-shot)	60.93
Winogrande (5-shot)	79.48
GSM8k (5-shot)	69.83