LazyMergekit-Qwen2.5-0.5B-Mixtral

LazyMergekit-Qwen2.5-0.5B-Mixtral is a Mixture of Experts (MoE) made with the following models using LazyMergekit:

• Qwen/Qwen2.5-0.5B-Instruct
• Qwen/Qwen2.5-Coder-0.5B
• funnyPhani/Qwen-2.5-0.5B-MATH
• caelancooper/Qwen2.5-0.5B-business
• KingNish/Qwen2.5-0.5b-Test-ft

🧩 Configuration

base_model: Qwen/Qwen2.5-0.5B-Instruct  # Base model for shared layers
gate_mode: hidden  # Use hidden representations for router initialization
dtype: float16  # Data type for the merged model

experts:
  - source_model: Qwen/Qwen2.5-0.5B-Instruct
    positive_prompts:
      - "chat"
      - "assistant"
      - "tell me"
      - "explain"
      - "I want"
  - source_model: Qwen/Qwen2.5-Coder-0.5B
    positive_prompts:
      - "code"
      - "python"
      - "javascript"
      - "programming"
      - "algorithm"
  - source_model: funnyPhani/Qwen-2.5-0.5B-MATH
    positive_prompts:
      - "math"
      - "mathematics"
      - "solve"
      - "count"
      - "reason"
  - source_model: caelancooper/Qwen2.5-0.5B-business
    positive_prompts:
      - "business"
      - "finance"
      - "market"
      - "strategy"
      - "analysis"
  - source_model: KingNish/Qwen2.5-0.5b-Test-ft
    positive_prompts:
      

💻 Usage

!pip install -qU transformers bitsandbytes accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Xiaojian9992024/LazyMergekit-Qwen2.5-0.5B-Mixtral"

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"])