Views
No views yet
UltraChat dataset, which contains a diverse range of synthetic dialogues generated by ChatGPT.
We then further aligned the model with 🤗 TRL's DPOTrainer on the openbmb/UltraFeedback dataset, which contain 64k prompts and model completions that are ranked by GPT-4."1# Install transformers from source - only needed for versions <= v4.34
2# pip install git+https://github.com/huggingface/transformers.git
3# pip install accelerate
4import torch
5from transformers import pipeline
6pipe = pipeline("text-generation", model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", torch_dtype=torch.bfloat16, device_map="auto")
7# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
8messages = [
9 {
10 "role": "system",
11 "content": "You are a friendly chatbot who always responds in the style of a pirate",
12 },
13 {"role": "user", "content": "How many helicopters can a human eat in one sitting?"},
14]
15prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
16outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
17print(outputs[0]["generated_text"])
18# <|system|>
19# You are a friendly chatbot who always responds in the style of a pirate.</s>
20# <|user|>
21# How many helicopters can a human eat in one sitting?</s>
22# <|assistant|>
23# ...