metadata

language:
  - en
  - hi
  - de
  - fr
  - ar
  - ja
license: apache-2.0
tags:
  - moe
model-index:
  - name: MetaModel_moe_multilingualv1
    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: 67.58
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=gagan3012/MetaModel_moe_multilingualv1
          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: 84.72
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=gagan3012/MetaModel_moe_multilingualv1
          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: 63.77
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=gagan3012/MetaModel_moe_multilingualv1
          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: 61.21
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=gagan3012/MetaModel_moe_multilingualv1
          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: 77.35
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=gagan3012/MetaModel_moe_multilingualv1
          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: 61.33
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=gagan3012/MetaModel_moe_multilingualv1
          name: Open LLM Leaderboard

MetaModel_moe_multilingualv1

This model is a Mixure of Experts (MoE) made with mergekit (mixtral branch). It uses the following base models:

🧩 Configuration

base_model: mlabonne/Marcoro14-7B-slerp
dtype: bfloat16
experts:
- positive_prompts:
  - chat
  - assistant
  - tell me
  - explain
  source_model: openchat/openchat-3.5-1210
- positive_prompts:
  - code
  - python
  - javascript
  - programming
  - algorithm
  source_model: beowolx/CodeNinja-1.0-OpenChat-7B
- positive_prompts:
  - storywriting
  - write
  - scene
  - story
  - character
  source_model: maywell/PiVoT-0.1-Starling-LM-RP
- positive_prompts:
  - reason
  - math
  - mathematics
  - solve
  - count
  source_model: WizardLM/WizardMath-7B-V1.1
- positive_prompts:
  - korean
  - answer in korean
  - korea
  source_model: davidkim205/komt-mistral-7b-v1
- positive_prompts:
  - chinese
  - china
  - answer in chinese
  source_model: OpenBuddy/openbuddy-zephyr-7b-v14.1
- positive_prompts:
  - hindi
  - india
  - hindu
  - answer in hindi
  source_model: manishiitg/open-aditi-hi-v1
- positive_prompts:
  - german
  - germany
  - answer in german
  - deutsch
  source_model: VAGOsolutions/SauerkrautLM-7b-v1-mistral
gate_mode: hidden

💻 Usage

!pip install -qU transformers bitsandbytes accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "gagan3012/MetaModel_moe_multilingualv1"

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.	69.33
AI2 Reasoning Challenge (25-Shot)	67.58
HellaSwag (10-Shot)	84.72
MMLU (5-Shot)	63.77
TruthfulQA (0-shot)	61.21
Winogrande (5-shot)	77.35
GSM8k (5-shot)	61.33