Hugging Face's logo

internlm
/
internlm2-20b

Text Generation
Transformers
PyTorch
internlm2
custom_code
Model card Files Community
5
internlm2-20b / README.md
leaderboard-pr-bot's picture
leaderboard-pr-bot
Adding Evaluation Results
d09c6c4 verified
preview code
|
raw
history blame
14.2 kB
metadata
license: other
pipeline_tag: text-generation
model-index:
  - name: internlm2-20b
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: AI2 Reasoning Challenge (25-Shot)
          type: ai2_arc
          config: ARC-Challenge
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: acc_norm
            value: 62.97
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=internlm/internlm2-20b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HellaSwag (10-Shot)
          type: hellaswag
          split: validation
          args:
            num_few_shot: 10
        metrics:
          - type: acc_norm
            value: 83.21
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=internlm/internlm2-20b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU (5-Shot)
          type: cais/mmlu
          config: all
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 67.58
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=internlm/internlm2-20b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: TruthfulQA (0-shot)
          type: truthful_qa
          config: multiple_choice
          split: validation
          args:
            num_few_shot: 0
        metrics:
          - type: mc2
            value: 51.27
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=internlm/internlm2-20b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Winogrande (5-shot)
          type: winogrande
          config: winogrande_xl
          split: validation
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 85.56
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=internlm/internlm2-20b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GSM8k (5-shot)
          type: gsm8k
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 67.93
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=internlm/internlm2-20b
          name: Open LLM Leaderboard

InternLM

Introduction

The second generation of the InternLM model, InternLM2, includes models at two scales: 7B and 20B. For the convenience of users and researchers, we have open-sourced four versions of each scale of the model, which are:

  • internlm2-base: A high-quality and highly adaptable model base, serving as an excellent starting point for deep domain adaptation.
  • internlm2 (recommended): Built upon the internlm2-base, this version has further pretrained on domain-specific corpus. It shows outstanding performance in evaluations while maintaining robust general language abilities, making it our recommended choice for most applications.
  • internlm2-chat-sft: Based on the Base model, it undergoes supervised human alignment training.
  • internlm2-chat (recommended): Optimized for conversational interaction on top of the internlm2-chat-sft through RLHF, it excels in instruction adherence, empathetic chatting, and tool invocation.

The base model of InternLM2 has the following technical features:

  • Effective support for ultra-long contexts of up to 200,000 characters: The model nearly perfectly achieves "finding a needle in a haystack" in long inputs of 200,000 characters. It also leads among open-source models in performance on long-text tasks such as LongBench and L-Eval.
  • Comprehensive performance enhancement: Compared to the previous generation model, it shows significant improvements in various capabilities, including reasoning, mathematics, and coding.

InternLM2-20B

Performance Evaluation

We have evaluated InternLM2 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.

Dataset\Models InternLM2-7B InternLM2-Chat-7B InternLM2-20B InternLM2-Chat-20B ChatGPT GPT-4
MMLU 65.8 63.7 67.7 66.5 69.1 83.0
AGIEval 49.9 47.2 53.0 50.3 39.9 55.1
BBH 65.0 61.2 72.1 68.3 70.1 86.7
GSM8K 70.8 70.7 76.1 79.6 78.2 91.4
MATH 20.2 23.0 25.5 31.9 28.0 45.8
HumanEval 43.3 59.8 48.8 67.1 73.2 74.4
MBPP(Sanitized) 51.8 51.4 63.0 65.8 78.9 79.0
  • 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 InternLM2-20B model using Transformers, use the following code:

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-20b", 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/internlm2-20b", 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)
# A beautiful flower with a long history of use in Ayurveda and traditional Chinese medicine. Known for its ability to help the body adapt to stress, it is a calming and soothing herb. It is used for its ability to help promote healthy sleep patterns, calm the nervous system and to help the body adapt to stress. It is also used for its ability to help the body deal with the symptoms of anxiety and depression. It is also used for its ability to help the body adapt to stress. It is also used for its ability to help the body adapt to stress. It is also used for its ability to help the

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.

简介

第二代浦语模型， InternLM2 包含 7B 和 20B 两个量级的模型。为了方便用户使用和研究，每个量级的模型我们总共开源了四个版本的模型，他们分别是

  • internlm2-base: 高质量和具有很强可塑性的模型基座，是模型进行深度领域适配的高质量起点；
  • internlm2（推荐）: 在internlm2-base基础上，进一步在特定领域的语料上进行预训练，在评测中成绩优异，同时保持了很好的通用语言能力，是我们推荐的在大部分应用中考虑选用的优秀基座；
  • internlm2-chat-sft：在Base基础上，进行有监督的人类对齐训练；
  • internlm2-chat（推荐）：在internlm2-chat-sft基础上，经过RLHF，面向对话交互进行了优化，具有很好的指令遵循、共情聊天和调用工具等的能力。

InternLM2 的基础模型具备以下的技术特点

  • 有效支持20万字超长上下文：模型在20万字长输入中几乎完美地实现长文“大海捞针”，而且在 LongBench 和 L-Eval 等长文任务中的表现也达到开源模型中的领先水平。
  • 综合性能全面提升：各能力维度相比上一代模型全面进步，在推理、数学、代码等方面的能力提升显著。

InternLM2-20B

性能评测

我们使用开源评测工具 OpenCompass 对 InternLM2 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。

评测集 InternLM2-7B InternLM2-Chat-7B InternLM2-20B InternLM2-Chat-20B ChatGPT GPT-4
MMLU 65.8 63.7 67.7 66.5 69.1 83.0
AGIEval 49.9 47.2 53.0 50.3 39.9 55.1
BBH 65.0 61.2 72.1 68.3 70.1 86.7
GSM8K 70.8 70.7 76.1 79.6 78.2 91.4
MATH 20.2 23.0 25.5 31.9 28.0 45.8
HumanEval 43.3 59.8 48.8 67.1 73.2 74.4
MBPP(Sanitized) 51.8 51.4 63.0 65.8 78.9 79.0
  • 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 InternLM2-20B 模型进行文本续写

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-20b", 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/internlm2-20b", 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)
# 来到美丽的大自然，我们欣赏着大自然的美丽风景，感受着大自然的气息。
# 今天，我来到了美丽的龙湾公园，这里风景秀丽，山清水秀，鸟语花香。一走进公园，我就被眼前的景象惊呆了：绿油油的草坪上，五颜六色的花朵竞相开放，散发出阵阵清香。微风吹来，花儿随风摆动，好像在向我们点头微笑。远处，巍峨的大山连绵起伏，好像一条巨龙在空中飞舞。山下，一条清澈的小河静静地流淌着，河里的鱼儿自由自在地

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。

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 69.75
AI2 Reasoning Challenge (25-Shot) 62.97
HellaSwag (10-Shot) 83.21
MMLU (5-Shot) 67.58
TruthfulQA (0-shot) 51.27
Winogrande (5-shot) 85.56
GSM8k (5-shot) 67.93