InternLM

  InternLM ^HOT  

💻Github Repo • 🤔Reporting Issues • 📜Technical Report

Introduction

While maintaining the InternLM2 architecture, various new technical explorations have been conducted, resulting in the next-generation model, InternLM2.5. InternLM2.5 leverages a large amount of synthetic data and continuously uses the InternLM for the iterative process of the model, thanks to the model's capability flywheel. The base model of InternLM2.5 has the following technical characteristics:

• Outstanding reasoning capability: Compared to InternLM2, the reasoning performance of InternLM2.5 has been improved by 20%;
• 1M Context window: Achieving near full accuracy in "finding a needle in a haystack" within a 1M context, and reaching leading levels among models of the same scale on LongBench and L-Eval.

InternLM2.5-20B

Performance Evaluation

We have evaluated InternLM2.5 on several important benchmarks using the open-source evaluation tool OpenCompass. Some of the evaluation results are shown in the table below. You are welcome to visit the OpenCompass Leaderboard for more evaluation results.

• The evaluation results were obtained from OpenCompass , and evaluation configuration can be found in the configuration files provided by OpenCompass.
• The evaluation data may have numerical differences due to the version iteration of OpenCompass, so please refer to the latest evaluation results of OpenCompass.

Limitations: Although we have made efforts to ensure the safety of the model during the training process and to encourage the model to generate text that complies with ethical and legal requirements, the model may still produce unexpected outputs due to its size and probabilistic generation paradigm. For example, the generated responses may contain biases, discrimination, or other harmful content. Please do not propagate such content. We are not responsible for any consequences resulting from the dissemination of harmful information.

Import from Transformers

To load the InternLM2.5-20B model using Transformers, use the following code:

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2_5-20b", trust_remote_code=True)
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
model = AutoModelForCausalLM.from_pretrained("internlm/internlm2_5-20b", torch_dtype=torch.float16, trust_remote_code=True).cuda()
model = model.eval()
inputs = tokenizer(["A beautiful flower"], return_tensors="pt")
for k,v in inputs.items():
    inputs[k] = v.cuda()
gen_kwargs = {"max_length": 128, "top_p": 0.8, "temperature": 0.8, "do_sample": True, "repetition_penalty": 1.0}
output = model.generate(**inputs, **gen_kwargs)
output = tokenizer.decode(output[0].tolist(), skip_special_tokens=True)
print(output)

Open Source License

The code is licensed under Apache-2.0, while model weights are fully open for academic research and also allow free commercial usage. To apply for a commercial license, please fill in the application form (English)/申请表（中文）. For other questions or collaborations, please contact internlm@pjlab.org.cn.

Citation

@misc{cai2024internlm2,
      title={InternLM2 Technical Report},
      author={Zheng Cai and Maosong Cao and Haojiong Chen and Kai Chen and Keyu Chen and Xin Chen and Xun Chen and Zehui Chen and Zhi Chen and Pei Chu and Xiaoyi Dong and Haodong Duan and Qi Fan and Zhaoye Fei and Yang Gao and Jiaye Ge and Chenya Gu and Yuzhe Gu and Tao Gui and Aijia Guo and Qipeng Guo and Conghui He and Yingfan Hu and Ting Huang and Tao Jiang and Penglong Jiao and Zhenjiang Jin and Zhikai Lei and Jiaxing Li and Jingwen Li and Linyang Li and Shuaibin Li and Wei Li and Yining Li and Hongwei Liu and Jiangning Liu and Jiawei Hong and Kaiwen Liu and Kuikun Liu and Xiaoran Liu and Chengqi Lv and Haijun Lv and Kai Lv and Li Ma and Runyuan Ma and Zerun Ma and Wenchang Ning and Linke Ouyang and Jiantao Qiu and Yuan Qu and Fukai Shang and Yunfan Shao and Demin Song and Zifan Song and Zhihao Sui and Peng Sun and Yu Sun and Huanze Tang and Bin Wang and Guoteng Wang and Jiaqi Wang and Jiayu Wang and Rui Wang and Yudong Wang and Ziyi Wang and Xingjian Wei and Qizhen Weng and Fan Wu and Yingtong Xiong and Chao Xu and Ruiliang Xu and Hang Yan and Yirong Yan and Xiaogui Yang and Haochen Ye and Huaiyuan Ying and Jia Yu and Jing Yu and Yuhang Zang and Chuyu Zhang and Li Zhang and Pan Zhang and Peng Zhang and Ruijie Zhang and Shuo Zhang and Songyang Zhang and Wenjian Zhang and Wenwei Zhang and Xingcheng Zhang and Xinyue Zhang and Hui Zhao and Qian Zhao and Xiaomeng Zhao and Fengzhe Zhou and Zaida Zhou and Jingming Zhuo and Yicheng Zou and Xipeng Qiu and Yu Qiao and Dahua Lin},
      year={2024},
      eprint={2403.17297},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

简介

在保持 InternLM2 架构的同时进行了多方面新的技术探索，得到新一代模型 InternLM2.5 ，新一代 InternLM2.5 利用了大量合成数据，并借助模型能力飞轮不断将 InternLM 用于模型的迭代过程。InternLM2.5 的基础模型具有以下的技术特点：

• 推理提升20%：相较于 InternLM2，InternLM2.5的推理性能提升了20%；
• 1M上下文：有效支持100万字超长上下文，在1M上下文的 “大海捞针” 实现全绿，并且在LongBench和L-Eval达到同量级模型领先水平。

InternLM2.5-20B

性能评测

我们使用开源评测工具 OpenCompass 对 InternLM2.5 在几个重要的评测集进行了评测 ，部分评测结果如下表所示，欢迎访问 OpenCompass 榜单 获取更多的评测结果。

• 以上评测结果基于 OpenCompass 获得（部分数据标注*代表数据来自原始论文），具体测试细节可参见 OpenCompass 中提供的配置文件。
• 评测数据会因 OpenCompass 的版本迭代而存在数值差异，请以 OpenCompass 最新版的评测结果为主。

局限性： 尽管在训练过程中我们非常注重模型的安全性，尽力促使模型输出符合伦理和法律要求的文本，但受限于模型大小以及概率生成范式，模型可能会产生各种不符合预期的输出，例如回复内容包含偏见、歧视等有害内容，请勿传播这些内容。由于传播不良信息导致的任何后果，本项目不承担责任。

通过 Transformers 加载

通过以下的代码加载 InternLM2.5-20B 模型进行文本续写

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2_5-20b", trust_remote_code=True)
# `torch_dtype=torch.float16` 可以令模型以 float16 精度加载，否则 transformers 会将模型加载为 float32，有可能导致显存不足
model = AutoModelForCausalLM.from_pretrained("internlm/internlm2_5-20b", torch_dtype=torch.float16, trust_remote_code=True).cuda()
model = model.eval()
inputs = tokenizer(["来到美丽的大自然"], return_tensors="pt")
for k,v in inputs.items():
    inputs[k] = v.cuda()
gen_kwargs = {"max_length": 128, "top_p": 0.8, "temperature": 0.8, "do_sample": True, "repetition_penalty": 1.0}
output = model.generate(**inputs, **gen_kwargs)
output = tokenizer.decode(output[0].tolist(), skip_special_tokens=True)
print(output)

开源许可证

本仓库的代码依照 Apache-2.0 协议开源。模型权重对学术研究完全开放，也可申请免费的商业使用授权（申请表）。其他问题与合作请联系 internlm@pjlab.org.cn。

引用

@misc{cai2024internlm2,
      title={InternLM2 Technical Report},
      author={Zheng Cai and Maosong Cao and Haojiong Chen and Kai Chen and Keyu Chen and Xin Chen and Xun Chen and Zehui Chen and Zhi Chen and Pei Chu and Xiaoyi Dong and Haodong Duan and Qi Fan and Zhaoye Fei and Yang Gao and Jiaye Ge and Chenya Gu and Yuzhe Gu and Tao Gui and Aijia Guo and Qipeng Guo and Conghui He and Yingfan Hu and Ting Huang and Tao Jiang and Penglong Jiao and Zhenjiang Jin and Zhikai Lei and Jiaxing Li and Jingwen Li and Linyang Li and Shuaibin Li and Wei Li and Yining Li and Hongwei Liu and Jiangning Liu and Jiawei Hong and Kaiwen Liu and Kuikun Liu and Xiaoran Liu and Chengqi Lv and Haijun Lv and Kai Lv and Li Ma and Runyuan Ma and Zerun Ma and Wenchang Ning and Linke Ouyang and Jiantao Qiu and Yuan Qu and Fukai Shang and Yunfan Shao and Demin Song and Zifan Song and Zhihao Sui and Peng Sun and Yu Sun and Huanze Tang and Bin Wang and Guoteng Wang and Jiaqi Wang and Jiayu Wang and Rui Wang and Yudong Wang and Ziyi Wang and Xingjian Wei and Qizhen Weng and Fan Wu and Yingtong Xiong and Chao Xu and Ruiliang Xu and Hang Yan and Yirong Yan and Xiaogui Yang and Haochen Ye and Huaiyuan Ying and Jia Yu and Jing Yu and Yuhang Zang and Chuyu Zhang and Li Zhang and Pan Zhang and Peng Zhang and Ruijie Zhang and Shuo Zhang and Songyang Zhang and Wenjian Zhang and Wenwei Zhang and Xingcheng Zhang and Xinyue Zhang and Hui Zhao and Qian Zhao and Xiaomeng Zhao and Fengzhe Zhou and Zaida Zhou and Jingming Zhuo and Yicheng Zou and Xipeng Qiu and Yu Qiao and Dahua Lin},
      year={2024},
      eprint={2403.17297},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}