CrystalMistral-26b is a Mixure of Experts (MoE) made with the following models using
LazyMergekit:
1base_model: Crystalcareai/CrystalMistral-13b
2gate_mode: random
3dtype: bfloat16 # output dtype (float32, float16, or bfloat16)
4experts_per_token: 2
5experts:
6 - source_model: Crystalcareai/CrystalMistral
7 positive_prompts:
8 - You are an helpful general-purpose assistant"
9 - source_model: Crystalcareai/CrystalMistral
10 positive_prompts:
11 - "You are an expert in providing detailed technical explanations."
12 - source_model: Crystalcareai/CrystalMistral
13 positive_prompts:
14 - "You are an expert in providing detailed code."
15 - source_model: Crystalcareai/CrystalMistral
16 positive_prompts:
17 - "You are an expert in providing emotionally emotional support."
18
1!pip install -qU transformers bitsandbytes accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "Crystalcareai/CrystalMistral-26b"
8
9tokenizer = AutoTokenizer.from_pretrained(model)
10pipeline = transformers.pipeline(
11 "text-generation",
12 model=model,
13 model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
14)
15
16messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
17prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
18outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
19print(outputs[0]["generated_text"])
