README.md

metadata

language:
  - en
license: cc-by-nc-4.0
tags:
  - merge
base_model:
  - janai-hq/trinity-v1
  - EmbeddedLLM/Mistral-7B-Merge-14-v0
model-index:
  - name: Mistral-7B-Merge-14-v0.2
    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: 68.86
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.2
          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: 87.01
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.2
          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: 65.05
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.2
          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: 64.19
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.2
          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: 81.69
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.2
          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: 70.51
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.2
          name: Open LLM Leaderboard

Update 2023-12-19

In light of dataset contamination issue among the merged models raised by the community in recent days, in particular berkeley-nest/Starling-LM-7B-alpha, Q-bert/MetaMath-Cybertron-Starling, and janai-hq/trinity-v1, we decided to remake another model without the models mentioned. Additionally, their CC-by-NC-4.0 license is restrictive and thus are not suitable for an open model.

Open LLM Leaderboard

For reference, this model obtained an average score of 72.88.

Average	72.88
ARC	68.86
HellaSwag	87.01
MMLU	65.05
TruthfulQA	64.19
Winogrande	81.69
GSM8K	70.51

Model Description

This is an experiment to test merging 14 models using DARE TIES 🦙

The merged model is then merged again with janai-hq/trinity-v1 using Gradient SLERP. The result is a base model that performs quite well but requires some further instruction fine-tuning.

The 14 models are as follows:

1. mistralai/Mistral-7B-Instruct-v0.2
2. ehartford/dolphin-2.2.1-mistral-7b
3. SciPhi/SciPhi-Mistral-7B-32k
4. ehartford/samantha-1.2-mistral-7b
5. Arc53/docsgpt-7b-mistral
6. berkeley-nest/Starling-LM-7B-alpha
7. Q-bert/MetaMath-Cybertron-Starling
8. Open-Orca/Mistral-7B-OpenOrca
9. v1olet/v1olet_marcoroni-go-bruins-merge-7B
10. beowolx/MistralHermes-CodePro-7B-v1
11. TIGER-Lab/MAmmoTH-7B-Mistral
12. teknium/OpenHermes-2.5-Mistral-7B
13. Weyaxi/OpenHermes-2.5-neural-chat-v3-3-Slerp
14. mlabonne/NeuralHermes-2.5-Mistral-7B

• base model: mistralai/Mistral-7B-v0.1

The yaml config file for this model is here:

slices:
  - sources:
      - model: janai-hq/trinity-v1
        layer_range: [0, 32]
      - model: EmbeddedLLM/Mistral-7B-Merge-14-v0
        layer_range: [0, 32]
merge_method: slerp
base_model: janai-hq/trinity-v1
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

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	72.88
AI2 Reasoning Challenge (25-Shot)	68.86
HellaSwag (10-Shot)	87.01
MMLU (5-Shot)	65.05
TruthfulQA (0-shot)	64.19
Winogrande (5-shot)	81.69
GSM8k (5-shot)	70.51