metadata
license: apache-2.0
tags:
- moe
- merge
- mergekit
- lazymergekit
- OpenPipe/mistral-ft-optimized-1227
- openchat/openchat-3.5-1210
- HuggingFaceH4/zephyr-7b-beta
- meta-math/MetaMath-Mistral-7B
OpenMistral-MoE
OpenMistral-MoE is a Mixure of Experts (MoE) made with the following models using LazyMergekit:
- OpenPipe/mistral-ft-optimized-1227
- openchat/openchat-3.5-1210
- HuggingFaceH4/zephyr-7b-beta
- meta-math/MetaMath-Mistral-7B
🧩 Configuration
base_model: mistralai/Mistral-7B-Instruct-v0.2
gate_mode: hidden
dtype: bfloat16
merge_method: dare_ties
experts:
- source_model: OpenPipe/mistral-ft-optimized-1227
positive_prompts:
- "chat"
- "assistant"
- "tell me"
- "explain"
- source_model: openchat/openchat-3.5-1210
positive_prompts:
- "code"
- "python"
- "javascript"
- "programming"
- "algorithm"
- source_model: HuggingFaceH4/zephyr-7b-beta
positive_prompts:
- "storywriting"
- "write"
- "scene"
- "story"
- "character"
- source_model: meta-math/MetaMath-Mistral-7B
positive_prompts:
- "reason"
- "math"
- "mathematics"
- "solve"
- "count"
💻 Usage
!pip install -qU transformers bitsandbytes accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Yash21/OpenMistral-MoE"
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"])