README.md

metadata

license: apache-2.0
tags:
  - moe
  - frankenmoe
  - merge
  - mergekit
  - lazymergekit
  - 222gate/Blurdus-7b-v0.1
  - 222gate/Blurred-Beagle-7b-slerp
  - liminerity/Blur-7b-v1.21
  - liminerity/Blur-7B-slerp-v0.1
base_model:
  - 222gate/Blurdus-7b-v0.1
  - 222gate/Blurred-Beagle-7b-slerp
  - liminerity/Blur-7b-v1.21
  - liminerity/Blur-7B-slerp-v0.1
model-index:
  - name: Blur-4x7b-MOE-v0.1
    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: 72.27
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=222gate/Blur-4x7b-MOE-v0.1
          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: 88.14
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=222gate/Blur-4x7b-MOE-v0.1
          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: 65.05
            name: accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=222gate/Blur-4x7b-MOE-v0.1
          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: 68.82
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=222gate/Blur-4x7b-MOE-v0.1
          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: 82.56
            name: accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=222gate/Blur-4x7b-MOE-v0.1
          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: 68.92
            name: accuracy
        source:
          url: >-
            https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=222gate/Blur-4x7b-MOE-v0.1
          name: Open LLM Leaderboard

Blur-4x7b-MOE-v0.1

Blur-4x7b-MOE-v0.1 is a Mixure of Experts (MoE) made with the following models using LazyMergekit:

• 222gate/Blurdus-7b-v0.1
• 222gate/Blurred-Beagle-7b-slerp
• liminerity/Blur-7b-v1.21
• liminerity/Blur-7B-slerp-v0.1

🧩 Configuration

base_model: 222gate/BrurryDog-7b-v0.1
gate_mode: hidden
dtype: bfloat16
experts:
  - source_model: 222gate/Blurdus-7b-v0.1
    positive_prompts:
    - "versatile"
    - "helpful"
    - "factual"
    - "integrated"
    - "adaptive"
    - "comprehensive"
    - "balanced"
    negative_prompts:
    - "specialized"
    - "narrow"
    - "focused"
    - "limited"
    - "specific"

  - source_model: 222gate/Blurred-Beagle-7b-slerp
    positive_prompts:
    - "creative"
    - "chat"
    - "discuss"
    - "culture"
    - "world"
    - "expressive"
    - "detailed"
    - "imaginative"
    - "engaging"
    negative_prompts:
    - "sorry"
    - "cannot"
    - "factual"
    - "concise"
    - "straightforward"
    - "objective"
    - "dry"

  - source_model: liminerity/Blur-7b-v1.21
    positive_prompts:
    - "analytical"
    - "accurate"
    - "logical"
    - "knowledgeable"
    - "precise"
    - "calculate"
    - "compute"
    - "solve"
    - "work"
    - "python"
    - "javascript"
    - "programming"
    - "algorithm"
    - "tell me"
    - "assistant"
    negative_prompts:
    - "creative"
    - "abstract"
    - "imaginative"
    - "artistic"
    - "emotional"
    - "mistake"
    - "inaccurate"

  - source_model: liminerity/Blur-7B-slerp-v0.1
    positive_prompts:
    - "instructive"
    - "clear"
    - "directive"
    - "helpful"
    - "informative"
    negative_prompts:
    - "exploratory"
    - "open-ended"
    - "narrative"
    - "speculative"
    - "artistic"

💻 Usage

!pip install -qU transformers bitsandbytes accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "222gate/Blur-4x7b-MOE-v0.1"

tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)

messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	74.29
AI2 Reasoning Challenge (25-Shot)	72.27
HellaSwag (10-Shot)	88.14
MMLU (5-Shot)	65.05
TruthfulQA (0-shot)	68.82
Winogrande (5-shot)	82.56
GSM8k (5-shot)	68.92