README.md

metadata

language:
  - en
license: apache-2.0
model-index:
  - name: HamSter-0.1
    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: 46.93
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/HamSter-0.1
          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: 68.08
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/HamSter-0.1
          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: 43.03
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/HamSter-0.1
          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: 51.24
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/HamSter-0.1
          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: 61.88
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/HamSter-0.1
          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: 0
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PotatoOff/HamSter-0.1
          name: Open LLM Leaderboard

Meet HamSter-0.1 🐹

👋 Uncensored fine tune model roleplay focused of "mistralai/Mistral-7B-v0.2" and first model of the HamSter series. Made with the help of my team ConvexAI.

🚀 For optimal performance, I recommend using a detailed character card! (There is NSFW chub.ai) Check out Chub.ai for some character cards.

🤩 Uses the Llama2 prompt template with chat instructions.

🔥 Produce spicy content.

😄 -> Check out HamSter 0.2 latest model of the HamSter series. Check it out!

HamSter 0.1 Quants Discord Server

I had good results with these parameters:

> temperature: 0.8 <

> top_p: 0.75

> min_p: 0

> top_k: 0

> repetition_penalty: 1.05

BenchMarks on OpenLLM Leaderboard

More details: HamSter-0.1 OpenLLM BenchMarks

BenchMarks on Ayumi's LLM Role Play & ERP Ranking

More details: Ayumi's LLM RolePlay & ERP Rankin HamSter-0.1 GGUF version Q6_K

Have Fun

💖

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	45.19
AI2 Reasoning Challenge (25-Shot)	46.93
HellaSwag (10-Shot)	68.08
MMLU (5-Shot)	43.03
TruthfulQA (0-shot)	51.24
Winogrande (5-shot)	61.88
GSM8k (5-shot)	0.00