README.md

metadata

language:
  - en
license: apache-2.0
tags:
  - mistral
  - not-for-all-audiences
  - merge
pipeline_tag: text-generation
inference: false
model-index:
  - name: DarkSapling-7B-v2.0
    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: 64.16
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TeeZee/DarkSapling-7B-v2.0
          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: 85.1
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TeeZee/DarkSapling-7B-v2.0
          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: 64.37
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TeeZee/DarkSapling-7B-v2.0
          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: 52.21
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TeeZee/DarkSapling-7B-v2.0
          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: 78.61
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TeeZee/DarkSapling-7B-v2.0
          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: 45.41
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TeeZee/DarkSapling-7B-v2.0
          name: Open LLM Leaderboard

DarkSapling-7B-v2.0

Model Details

• A result of 4 models merge. DARE TIES method was used this time so resulting model better preserves characteristics of all included models than v1.x.
• models used for merge: cognitivecomputations/dolphin-2.6-mistral-7b-dpo-laser KoboldAI/Mistral-7B-Holodeck-1 KoboldAI/Mistral-7B-Erebus-v3 cognitivecomputations/samantha-mistral-7b
• See mergekit-config.yml for details on the merge method used.

Warning: This model can produce NSFW content!

Results

• a little different than version v1.0, more romantic and empathetic.
• smarter than versions 1.0 and 1.1.
• best for one-on-one ERP.
• produces SFW nad NSFW content without issues, switches context seamlessly.
• sticks to character card
• pretty smart due to mistral, empathetic after Samantha and sometimes produces dark scenarions - Erebus.
• storytelling is satisfactory due to Holodeck
• good at following instructions

All comments are greatly appreciated, download, test and if you appreciate my work, consider buying me my fuel:

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	64.98
AI2 Reasoning Challenge (25-Shot)	64.16
HellaSwag (10-Shot)	85.10
MMLU (5-Shot)	64.37
TruthfulQA (0-shot)	52.21
Winogrande (5-shot)	78.61
GSM8k (5-shot)	45.41