Views
No views yet
1!pip install -qU transformers bitsandbytes accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "Isotonic/smol_llama-4x220M-MoE"
8
9tokenizer = AutoTokenizer.from_pretrained(model)
10pipeline = transformers.pipeline(
11 "text-generation",
12 model=model,
13 model_kwargs={"torch_dtype": torch.bfloat16},
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"])1experts:
2 - source_model: BEE-spoke-data/smol_llama-220M-openhermes
3 positive_prompts:
4 - "reasoning"
5 - "logic"
6 - "problem-solving"
7 - "critical thinking"
8 - "analysis"
9 - "synthesis"
10 - "evaluation"
11 - "decision-making"
12 - "judgment"
13 - "insight"
14
15 - source_model: BEE-spoke-data/beecoder-220M-python
16 positive_prompts:
17 - "program"
18 - "software"
19 - "develop"
20 - "build"
21 - "create"
22 - "design"
23 - "implement"
24 - "debug"
25 - "test"
26 - "code"
27 - "python"
28 - "programming"
29 - "algorithm"
30 - "function"
31
32 - source_model: BEE-spoke-data/zephyr-220m-sft-full
33 positive_prompts:
34 - "storytelling"
35 - "narrative"
36 - "fiction"
37 - "creative writing"
38 - "plot"
39 - "characters"
40 - "dialogue"
41 - "setting"
42 - "emotion"
43 - "imagination"
44 - "scene"
45 - "story"
46 - "character"
47
48 - source_model: BEE-spoke-data/zephyr-220m-dpo-full
49 positive_prompts:
50 - "chat"
51 - "conversation"
52 - "dialogue"
53 - "discuss"
54 - "ask questions"
55 - "share thoughts"
56 - "explore ideas"
57 - "learn new things"
58 - "personal assistant"
59 - "friendly helper"