Stablebeluga2

🚀 Stable Beluga 2

Stable Beluga 2是基于Llama2 70B模型微调的自回归语言模型，适用于文本生成任务。本仓库对原模型进行了优化，在存储和加载方面有更好的表现。

🚀 快速开始

使用以下代码片段开始与 Stable Beluga 2 进行对话：

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

tokenizer = AutoTokenizer.from_pretrained("stabilityai/StableBeluga2", use_fast=False)
model = AutoModelForCausalLM.from_pretrained("stabilityai/StableBeluga2", torch_dtype=torch.float16, low_cpu_mem_usage=True, device_map="auto")
system_prompt = "### System:\nYou are Stable Beluga, an AI that follows instructions extremely well. Help as much as you can. Remember, be safe, and don't do anything illegal.\n\n"

message = "Write me a poem please"
prompt = f"{system_prompt}### User: {message}\n\n### Assistant:\n"
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
output = model.generate(**inputs, do_sample=True, top_p=0.95, top_k=0, max_new_tokens=256)

print(tokenizer.decode(output[0], skip_special_tokens=True))

Stable Beluga 2 应使用以下提示格式：

### System:
This is a system prompt, please behave and help the user.

### User:
Your prompt here

### Assistant:
The output of Stable Beluga 2

✨ 主要特性

本仓库包含来自 stabilityai/StableBeluga2 仓库的模型，并进行了以下更改：

1. 以 bfloat16 而非 float32 存储权重：这使得文件大小缩小了2倍，同时会有轻微的质量损失，但与Petals默认使用的NF4量化所导致的损失相比，并不显著。
2. 以小分片形式存储权重：每个Transformer块存储在其自己的分片中（每个1.71 GB）。输入和输出嵌入以及相邻的层归一化也存储在单独的分片中（1.05 GB）。这样，Petals客户端和服务器无需下载除实际使用的层之外的任何多余数据。
3. 使用 Safetensors 而非Pickle：这允许更快的加载速度和更低的内存要求。

📚 详细文档

模型描述

Stable Beluga 2 是在Llama2 70B模型基础上，针对Orca风格数据集进行微调的模型。

模型详情

训练数据集

Stable Beluga 2 在内部的Orca风格数据集上进行训练。

训练过程

模型通过在上述数据集上进行监督微调进行学习，采用混合精度（BF16）训练，并使用AdamW进行优化。以下是相关超参数：

伦理考量与局限性

Beluga是一项新技术，使用时存在风险。到目前为止的测试仅使用英语进行，且无法涵盖所有场景。因此，与所有大语言模型一样，Beluga的潜在输出无法提前预测，在某些情况下，模型可能会对用户提示产生不准确、有偏见或其他令人反感的响应。因此，在部署Beluga的任何应用之前，开发人员应针对其特定应用对模型进行安全测试和调整。

引用方式

@misc{StableBelugaModels, 
      url={[https://huggingface.co/stabilityai/StableBeluga2](https://huggingface.co/stabilityai/StableBeluga2)}, 
      title={Stable Beluga models}, 
      author={Mahan, Dakota and Carlow, Ryan and Castricato, Louis and Cooper, Nathan and Laforte, Christian}
}

参考文献

@misc{touvron2023llama,
      title={Llama 2: Open Foundation and Fine-Tuned Chat Models}, 
      author={Hugo Touvron and Louis Martin and Kevin Stone and Peter Albert and Amjad Almahairi and Yasmine Babaei and Nikolay Bashlykov and Soumya Batra and Prajjwal Bhargava and Shruti Bhosale and Dan Bikel and Lukas Blecher and Cristian Canton Ferrer and Moya Chen and Guillem Cucurull and David Esiobu and Jude Fernandes and Jeremy Fu and Wenyin Fu and Brian Fuller and Cynthia Gao and Vedanuj Goswami and Naman Goyal and Anthony Hartshorn and Saghar Hosseini and Rui Hou and Hakan Inan and Marcin Kardas and Viktor Kerkez and Madian Khabsa and Isabel Kloumann and Artem Korenev and Punit Singh Koura and Marie-Anne Lachaux and Thibaut Lavril and Jenya Lee and Diana Liskovich and Yinghai Lu and Yuning Mao and Xavier Martinet and Todor Mihaylov and Pushkar Mishra and Igor Molybog and Yixin Nie and Andrew Poulton and Jeremy Reizenstein and Rashi Rungta and Kalyan Saladi and Alan Schelten and Ruan Silva and Eric Michael Smith and Ranjan Subramanian and Xiaoqing Ellen Tan and Binh Tang and Ross Taylor and Adina Williams and Jian Xiang Kuan and Puxin Xu and Zheng Yan and Iliyan Zarov and Yuchen Zhang and Angela Fan and Melanie Kambadur and Sharan Narang and Aurelien Rodriguez and Robert Stojnic and Sergey Edunov and Thomas Scialom},
      year={2023},
      eprint={2307.09288},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

@misc{mukherjee2023orca,
      title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4}, 
      author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
      year={2023},
      eprint={2306.02707},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

其他Beluga模型

• StableBeluga 1 - Delta
• StableBeluga 13B
• StableBeluga 7B