Noro-Hermes-3x7B
Noro-Hermes-3x7B is a Mixture of Experts (MoE) made with the following models using LazyMergekit:
- NousResearch/Nous-Hermes-2-Mistral-7B-DPO
- NeverSleep/Noromaid-7B-0.4-DPO
- mistralai/Mistral-7B-Instruct-v0.2
🧩 Configuration
base_model: mistralai/Mistral-7B-Instruct-v0.2
dtype: bfloat16
gate_mode: hidden
experts:
- source_model: NousResearch/Nous-Hermes-2-Mistral-7B-DPO
positive_prompts: ["You are an intelligent assistant."]
- source_model: NeverSleep/Noromaid-7B-0.4-DPO
positive_prompts: ["You are a creative roleplaying assistant."]
- source_model: mistralai/Mistral-7B-Instruct-v0.2
positive_prompts: ["You are a general purpose assistant."]
💻 Usage
!pip install -qU transformers bitsandbytes accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "ThomasComics/Noro-Hermes-3x7B"
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"])