CBT LLM

🚀 Instruction-tuned LoRA Model for Question Answering

This is an instruction-tuned LoRA model based on https://huggingface.co/baichuan-inc/baichuan-7B, designed for question-answering tasks.

🚀 Quick Start

Prerequisites

Ensure you follow the baichuan-7B License when using this model.

Installation

To install the necessary package, run the following command:

pip install transformers==4.33.2

Usage Examples

Basic Usage

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer

tokenizer = AutoTokenizer.from_pretrained("CBT-LLM", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("CBT-LLM", trust_remote_code=True).cuda()
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

query = "为什么生怕一点点事情做不好被人批评？做事情，别人告诉了我方法，但身体不会按方法来，非要折腾几遍，才发现别人告诉的方法和口诀，是最高效的；而且最近一个月，睡眠不好，整个白天都是无精打采的，每天活的很丧，知道自己的问题出在哪里，不晓得怎么去做出改变"
template = (
    "你是一名经验丰富的心理咨询师，专长于认知行为疗法, 以心理咨询师的身份回答以下问题。\n"
    "Human: {}\nAssistant: "
)

inputs = tokenizer([template.format(query)], return_tensors="pt")
inputs = inputs.to("cuda")
generate_ids = model.generate(**inputs, max_new_tokens=256, streamer=streamer)

📄 License

This project is licensed under the Apache-2.0 License.

Additional Information

• Training framework: https://github.com/hiyouga/LLaMA-Factory