github
/
InternLM
mirror of https://github.com/InternLM/InternLM
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

[Doc]: Update README (#599)

pull/601/head
Yang Gao 2024-01-17 13:07:35 +08:00 committed by GitHub
parent 8c546c9eb9
commit 896bac06a6
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
10 changed files with 74 additions and 66 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

54
README.md
View File

@ -58,22 +58,24 @@ InternLM2 series are released with the following features:
| Model | Transformers(HF) | ModelScope(HF) | OpenXLab(HF) | Release Date | | Model | Transformers(HF) | ModelScope(HF) | OpenXLab(HF) | Release Date |
|---------------------------|------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------| |---------------------------|------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------|
| **InternLM2 Chat 20B** | [🤗internlm/internlm2-chat-20b](https://huggingface.co/internlm/internlm2-chat-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-20b) | 2024-01-17 | | **InternLM2-Base-7B** | [🤗internlm/internlm2-base-7b](https://huggingface.co/internlm/internlm2-base-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-base-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-base-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-base-7b) | 2024-01-17 |
| **InternLM2 20B** | [🤗internlm/internlm2-20b](https://huggingface.co/internlm/internlm2-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-20b) | 2024-01-17 | | **InternLM2-7B** | [🤗internlm/internlm2-7b](https://huggingface.co/internlm/internlm2-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-7b) | 2024-01-17 |
| **InternLM2 Chat 20B SFT** | [🤗internlm/internlm2-chat-20b-sft](https://huggingface.co/internlm/internlm2-chat-20b-sft) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-20b-sft](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-20b-sft/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-20b-sft) | 2024-01-17 | | **InternLM2-Chat-7B-SFT** | [🤗internlm/internlm2-chat-7b-sft](https://huggingface.co/internlm/internlm2-chat-7b-sft) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-7b-sft](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-7b-sft/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-7b-sft) | 2024-01-17 |
| **InternLM2 Base 20B** | [🤗internlm/internlm2-base-20b](https://huggingface.co/internlm/internlm2-base-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-base-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-base-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-base-20b) | 2024-01-17 | | **InternLM2-Chat-7B** | [🤗internlm/internlm2-chat-7b](https://huggingface.co/internlm/internlm2-chat-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-7b) | 2024-01-17 |
| **InternLM2 Chat 7B** | [🤗internlm/internlm2-chat-7b](https://huggingface.co/internlm/internlm2-chat-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-7b) | 2024-01-17 | | **InternLM2-Base-20B** | [🤗internlm/internlm2-base-20b](https://huggingface.co/internlm/internlm2-base-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-base-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-base-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-base-20b) | 2024-01-17 |
| **InternLM2 7B** | [🤗internlm/internlm2-7b](https://huggingface.co/internlm/internlm2-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-7b) | 2024-01-17 | | **InternLM2-20B** | [🤗internlm/internlm2-20b](https://huggingface.co/internlm/internlm2-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-20b) | 2024-01-17 |
| **InternLM2 Chat 7B SFT** | [🤗internlm/internlm2-chat-7b-sft](https://huggingface.co/internlm/internlm2-chat-7b-sft) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-7b-sft](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-7b-sft/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-7b-sft) | 2024-01-17 | | **InternLM2-Chat-20B-SFT** | [🤗internlm/internlm2-chat-20b-sft](https://huggingface.co/internlm/internlm2-chat-20b-sft) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-20b-sft](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-20b-sft/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-20b-sft) | 2024-01-17 |
| **InternLM2 Base 7B** | [🤗internlm/internlm2-base-7b](https://huggingface.co/internlm/internlm2-base-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-base-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-base-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-base-7b) | 2024-01-17 | | **InternLM2-Chat-20B** | [🤗internlm/internlm2-chat-20b](https://huggingface.co/internlm/internlm2-chat-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-20b) | 2024-01-17 |
**Note of Models:** **Note of Models:**
The release of InternLM2 series contains two model sizes: 7B and 20B. 7B models are efficient for research and application and 20B models are more powerful and can support more complex scenarios. For each model size, there are three types of models for different user requirements The release of InternLM2 series contains two model sizes: 7B and 20B. 7B models are efficient for research and application and 20B models are more powerful and can support more complex scenarios. For each model size, there are four types of models for different user requirements
1. InternLM2-Base: Foundation models with high quality and high adaptation flexibility, which serve as a good starting point for downstream deep adaptations. 1. InternLM2-Base: Foundation models with high quality and high adaptation flexibility, which serve as a good starting point for downstream deep adaptations.
2. InternLM2: Optimized in multiple dimensions based on InternLM2-Base, obtaining state-of-the-art performance in evaluation with good language capability. InternLM2 models are recommended for consideration in most applications. 2. InternLM2: Optimized in multiple dimensions based on InternLM2-Base, obtaining state-of-the-art performance in evaluation with good language capability. InternLM2 models are recommended for consideration in most applications.
3. InternLM2-Chat: InternLM2-Chat have gone through SFT and online RLHF based on InternLM2-Base model, for better instruction following, chat experience and function calling, which is recommended for downstream applications. We also released their corresponding SFT version, termed InternLM2 Chat 7/20B SFT, to ease the research for alignment. 3. InternLM2-Chat-SFT: Based on the InternLM2-Base model, it undergoes supervised human alignment training.
3. InternLM2-Chat: Optimized for conversational interaction on top of the InternLM2-Chat-SFT through RLHF, it excels in instruction adherence, empathetic chatting, and tool invocation, for better instruction following, chat experience and function calling, which is recommended for downstream applications.
**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. **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.
@ -126,30 +128,33 @@ The chat models adopt [chatml format](./chat/chat_format.md) to support both cha
### Import from Transformers ### Import from Transformers
To load the InternLM2 7B Chat model using Transformers, use the following code: To load the InternLM2-7B-Chat model using Transformers, use the following code:
```python ```python
>>> from transformers import AutoTokenizer, AutoModelForCausalLM import torch
>>> tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-chat-7b", trust_remote_code=True) from transformers import AutoTokenizer, AutoModelForCausalLM
>>> model = AutoModelForCausalLM.from_pretrained("internlm/internlm2-chat-7b", trust_remote_code=True).cuda() tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-chat-7b", trust_remote_code=True)
>>> model = model.eval() # Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
>>> response, history = model.chat(tokenizer, "hello", history=[]) model = AutoModelForCausalLM.from_pretrained("internlm/internlm2-chat-7b", trust_remote_code=True, torch_dtype=torch.float16).cuda()
>>> print(response) model = model.eval()
Hello! How can I help you today? response, history = model.chat(tokenizer, "hello", history=[])
>>> response, history = model.chat(tokenizer, "please provide three suggestions about time management", history=history) print(response)
>>> print(response) # Output: Hello? How can I help you today?
response, history = model.chat(tokenizer, "please provide three suggestions about time management", history=history)
print(response)
``` ```
### Import from ModelScope ### Import from ModelScope
To load the InternLM model using ModelScope, use the following code: To load the InternLM2-7B-Chat model using ModelScope, use the following code:
```python ```python
from modelscope import snapshot_download, AutoTokenizer, AutoModelForCausalLM
import torch import torch
from modelscope import snapshot_download, AutoTokenizer, AutoModelForCausalLM
model_dir = snapshot_download('Shanghai_AI_Laboratory/internlm2-chat-7b') model_dir = snapshot_download('Shanghai_AI_Laboratory/internlm2-chat-7b')
tokenizer = AutoTokenizer.from_pretrained(model_dir, device_map="auto", trust_remote_code=True,torch_dtype=torch.float16) tokenizer = AutoTokenizer.from_pretrained(model_dir, device_map="auto", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_dir,device_map="auto", trust_remote_code=True,torch_dtype=torch.float16) # 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(model_dir, device_map="auto", trust_remote_code=True, torch_dtype=torch.float16)
model = model.eval() model = model.eval()
response, history = model.chat(tokenizer, "hello", history=[]) response, history = model.chat(tokenizer, "hello", history=[])
print(response) print(response)
@ -192,7 +197,6 @@ Please refer to [finetune docs](./finetune/) for fine-tuning with InternLM.
**Note:** We have migrated the whole training functionality in this project to [InternEvo](https://github.com/InternLM/InternEvo) for easier user experience, which provides efficient pre-training and fine-tuning infra for training InternLM. **Note:** We have migrated the whole training functionality in this project to [InternEvo](https://github.com/InternLM/InternEvo) for easier user experience, which provides efficient pre-training and fine-tuning infra for training InternLM.
## Evaluation ## Evaluation
We utilize [OpenCompass](https://github.com/open-compass/opencompass) for model evaluation. In InternLM-2, we primarily focus on standard objective evaluation, long-context evaluation (needle in a haystack), data contamination assessment, agent evaluation, and subjective evaluation. We utilize [OpenCompass](https://github.com/open-compass/opencompass) for model evaluation. In InternLM-2, we primarily focus on standard objective evaluation, long-context evaluation (needle in a haystack), data contamination assessment, agent evaluation, and subjective evaluation.

53
README_zh-CN.md
View File

@ -56,22 +56,23 @@ InternLM2 系列模型在本仓库正式发布，具有如下特性：
| Model | Transformers(HF) | ModelScope(HF) | OpenXLab(HF) | Release Date | | Model | Transformers(HF) | ModelScope(HF) | OpenXLab(HF) | Release Date |
|---------------------------|------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------| |---------------------------|------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------|
| **InternLM2 Chat 20B** | [🤗internlm/internlm2-chat-20b](https://huggingface.co/internlm/internlm2-chat-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-20b) | 2024-01-17 | | **InternLM2-Base-7B** | [🤗internlm/internlm2-base-7b](https://huggingface.co/internlm/internlm2-base-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-base-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-base-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-base-7b) | 2024-01-17 |
| **InternLM2 20B** | [🤗internlm/internlm2-20b](https://huggingface.co/internlm/internlm2-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-20b) | 2024-01-17 | | **InternLM2-7B** | [🤗internlm/internlm2-7b](https://huggingface.co/internlm/internlm2-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-7b) | 2024-01-17 |
| **InternLM2 Chat 20B SFT** | [🤗internlm/internlm2-chat-20b-sft](https://huggingface.co/internlm/internlm2-chat-20b-sft) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-20b-sft](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-20b-sft/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-20b-sft) | 2024-01-17 | | **InternLM2-Chat-7B-SFT** | [🤗internlm/internlm2-chat-7b-sft](https://huggingface.co/internlm/internlm2-chat-7b-sft) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-7b-sft](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-7b-sft/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-7b-sft) | 2024-01-17 |
| **InternLM2 Base 20B** | [🤗internlm/internlm2-base-20b](https://huggingface.co/internlm/internlm2-base-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-base-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-base-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-base-20b) | 2024-01-17 | | **InternLM2-Chat-7B** | [🤗internlm/internlm2-chat-7b](https://huggingface.co/internlm/internlm2-chat-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-7b) | 2024-01-17 |
| **InternLM2 Chat 7B** | [🤗internlm/internlm2-chat-7b](https://huggingface.co/internlm/internlm2-chat-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-7b) | 2024-01-17 | | **InternLM2-Base-20B** | [🤗internlm/internlm2-base-20b](https://huggingface.co/internlm/internlm2-base-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-base-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-base-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-base-20b) | 2024-01-17 |
| **InternLM2 7B** | [🤗internlm/internlm2-7b](https://huggingface.co/internlm/internlm2-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-7b) | 2024-01-17 | | **InternLM2-20B** | [🤗internlm/internlm2-20b](https://huggingface.co/internlm/internlm2-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-20b) | 2024-01-17 |
| **InternLM2 Chat 7B SFT** | [🤗internlm/internlm2-chat-7b-sft](https://huggingface.co/internlm/internlm2-chat-7b-sft) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-7b-sft](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-7b-sft/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-7b-sft) | 2024-01-17 | | **InternLM2-Chat-20B-SFT** | [🤗internlm/internlm2-chat-20b-sft](https://huggingface.co/internlm/internlm2-chat-20b-sft) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-20b-sft](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-20b-sft/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-20b-sft) | 2024-01-17 |
| **InternLM2 Base 7B** | [🤗internlm/internlm2-base-7b](https://huggingface.co/internlm/internlm2-base-7b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-base-7b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-base-7b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-base-7b) | 2024-01-17 | | **InternLM2-Chat-20B** | [🤗internlm/internlm2-chat-20b](https://huggingface.co/internlm/internlm2-chat-20b) | [<img src="./assets/modelscope_logo.png" width="20px" /> internlm2-chat-20b](https://modelscope.cn/models/Shanghai_AI_Laboratory/internlm2-chat-20b/summary) | [![Open in OpenXLab](https://cdn-static.openxlab.org.cn/header/openxlab_models.svg)](https://openxlab.org.cn/models/detail/OpenLMLab/internlm2-chat-20b) | 2024-01-17 |
**关于模型说明：** **关于模型说明：**
在此次发布中，InternLM2 包含两种模型规格：7B 和 20B。7B 为轻量级的研究和应用提供了一个轻便但性能不俗的模型，20B 模型的综合性能更为强劲，可以有效支持更加复杂的实用场景。面向不同的使用需求，每个规格包含三个模型版本： 在此次发布中，InternLM2 包含两种模型规格：7B 和 20B。7B 为轻量级的研究和应用提供了一个轻便但性能不俗的模型，20B 模型的综合性能更为强劲，可以有效支持更加复杂的实用场景。面向不同的使用需求，每个规格包含四个模型版本：
1. 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。 1. 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。
2. InternLM2：在 Base 模型基础上，在多个能力方向进行了强化，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。 2. InternLM2：在 Base 模型基础上，在多个能力方向进行了强化，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。
3. InternLM2-Chat：InternLM2-Chat 模型在 InternLM2-Base 模型的基础上，经过了 SFT 和 RLHF，面向对话交互进行了优化，具有较好的指令遵循、共情聊天和调用工具等的能力，是我们推荐直接用于下游应用的模型。我们同时开源了这些模型使用的 SFT 版本方便社区的对齐研究，标记为 InternLM2-Chat 7B/20B SFT。 3. InternLM2-Chat-SFT: 基于InternLM2-Base模型基座，进行有监督的微调对齐训练。
3. InternLM2-Chat：在 InternLM2-Chat-SFT 的基础上，通过基于人类反馈的强化学习算法进行了优化，以更好地适应对话交互，并在指令遵循、情感交流和功能调用方面表现出色，从而为下游应用提供更好的指令遵循、聊天体验和功能调用，这是我们推荐的在下游应用中的对话模型选择。
**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。 **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。
@ -124,18 +125,20 @@ InternLM2 系列模型在本仓库正式发布，具有如下特性：
### 通过 Transformers 加载 ### 通过 Transformers 加载
通过以下的代码从 Transformers 加载 InternLM 模型 （可修改模型名称替换不同的模型） 通过以下的代码从 Transformers 加载 InternLM2-7B-Chat 模型 （可修改模型名称替换不同的模型）
```python ```python
>>> from transformers import AutoTokenizer, AutoModelForCausalLM import torch
>>> tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-chat-7b", trust_remote_code=True) from transformers import AutoTokenizer, AutoModelForCausalLM
>>> model = AutoModelForCausalLM.from_pretrained("internlm/internlm2-chat-7b", trust_remote_code=True).cuda() tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-chat-7b", trust_remote_code=True)
>>> model = model.eval() # 设置`torch_dtype=torch.float16`来将模型精度指定为torch.float16，否则可能会因为您的硬件原因造成显存不足的问题。
>>> response, history = model.chat(tokenizer, "你好", history=[]) model = AutoModelForCausalLM.from_pretrained("internlm/internlm2-chat-7b", trust_remote_code=True, torch_dtype=torch.float16).cuda()
>>> print(response) model = model.eval()
你好！有什么我可以帮助你的吗？ response, history = model.chat(tokenizer, "你好", history=[])
>>> response, history = model.chat(tokenizer, "请提供三个管理时间的建议。", history=history) print(response)
>>> print(response) # 模型输出：你好！有什么我可以帮助你的吗？
response, history = model.chat(tokenizer, "请提供三个管理时间的建议。", history=history)
print(response)
``` ```
### 通过 ModelScope 加载 ### 通过 ModelScope 加载
@ -143,11 +146,11 @@ InternLM2 系列模型在本仓库正式发布，具有如下特性：
通过以下的代码从 ModelScope 加载 InternLM 模型 （可修改模型名称替换不同的模型） 通过以下的代码从 ModelScope 加载 InternLM 模型 （可修改模型名称替换不同的模型）
```python ```python
from modelscope import snapshot_download, AutoTokenizer, AutoModelForCausalLM
import torch import torch
from modelscope import snapshot_download, AutoTokenizer, AutoModelForCausalLM
model_dir = snapshot_download('Shanghai_AI_Laboratory/internlm2-chat-7b') model_dir = snapshot_download('Shanghai_AI_Laboratory/internlm2-chat-7b')
tokenizer = AutoTokenizer.from_pretrained(model_dir, device_map="auto", trust_remote_code=True,torch_dtype=torch.float16) tokenizer = AutoTokenizer.from_pretrained(model_dir, device_map="auto", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_dir,device_map="auto", trust_remote_code=True,torch_dtype=torch.float16) model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, torch_dtype=torch.float16)
model = model.eval() model = model.eval()
response, history = model.chat(tokenizer, "hello", history=[]) response, history = model.chat(tokenizer, "hello", history=[])
print(response) print(response)
@ -165,10 +168,6 @@ pip install transformers==4.30.2
streamlit run ./chat/web_demo.py streamlit run ./chat/web_demo.py
``` ```
效果如下
![效果](https://github.com/InternLM/InternLM/assets/9102141/11b60ee0-47e4-42c0-8278-3051b2f17fe4)
### 基于 InternLM 高性能部署 ### 基于 InternLM 高性能部署
我们使用 [LMDeploy](https://github.com/InternLM/LMDeploy) 完成 InternLM 的一键部署。 我们使用 [LMDeploy](https://github.com/InternLM/LMDeploy) 完成 InternLM 的一键部署。

2
assets/compass_support.svg
View File

@ -1 +1 @@
<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="159" height="20" role="img" aria-label="OpenCompass: Support"><title>OpenCompass: Support</title><linearGradient id="s" x2="0" y2="100%"><stop offset="0" stop-color="#bbb" stop-opacity=".1"/><stop offset="1" stop-opacity=".1"/></linearGradient><clipPath id="r"><rect width="159" height="20" rx="3" fill="#fff"/></clipPath><g clip-path="url(#r)"><rect width="106" height="20" fill="#555"/><rect x="106" width="53" height="20" fill="royalblue"/><rect width="159" height="20" fill="url(#s)"/></g><g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" text-rendering="geometricPrecision" font-size="110"><image x="5" y="3" width="14" height="14" xlink:href="data:image/svg+xml;base64,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"/><text aria-hidden="true" x="625" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="790">OpenCompass</text><text x="625" y="140" transform="scale(.1)" fill="#fff" textLength="790">OpenCompass</text><text aria-hidden="true" x="1315" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="430">Support</text><text x="1315" y="140" transform="scale(.1)" fill="#fff" textLength="430">Support</text></g></svg> <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="159" height="20" role="img" aria-label="OpenCompass: Support"><title>OpenCompass: Support</title><linearGradient id="s" x2="0" y2="100%"><stop offset="0" stop-color="#bbb" stop-opacity=".1"/><stop offset="1" stop-opacity=".1"/></linearGradient><clipPath id="r"><rect width="159" height="20" rx="3" fill="#fff"/></clipPath><g clip-path="url(#r)"><rect width="106" height="20" fill="#555"/><rect x="106" width="53" height="20" fill="royalblue"/><rect width="159" height="20" fill="url(#s)"/></g><g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" text-rendering="geometricPrecision" font-size="110"><image x="5" y="3" width="14" height="14" xlink:href="data:image/svg+xml;base64,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"/><text aria-hidden="true" x="625" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="790">OpenCompass</text><text x="625" y="140" transform="scale(.1)" fill="#fff" textLength="790">OpenCompass</text><text aria-hidden="true" x="1315" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="430">Support</text><text x="1315" y="140" transform="scale(.1)" fill="#fff" textLength="430">Support</text></g></svg>

Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 4.3 KiB

After

Width:  |  Height:  |  Size: 4.3 KiB

2
assets/license.svg
View File

@ -1 +1 @@
<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="120" height="20" role="img" aria-label="license: Apache-2.0"><title>license: Apache-2.0</title><linearGradient id="s" x2="0" y2="100%"><stop offset="0" stop-color="#bbb" stop-opacity=".1"/><stop offset="1" stop-opacity=".1"/></linearGradient><clipPath id="r"><rect width="120" height="20" rx="3" fill="#fff"/></clipPath><g clip-path="url(#r)"><rect width="47" height="20" fill="#555"/><rect x="47" width="73" height="20" fill="#97ca00"/><rect width="120" height="20" fill="url(#s)"/></g><g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" text-rendering="geometricPrecision" font-size="110"><text aria-hidden="true" x="245" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="370">license</text><text x="245" y="140" transform="scale(.1)" fill="#fff" textLength="370">license</text><text aria-hidden="true" x="825" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="630">Apache-2.0</text><text x="825" y="140" transform="scale(.1)" fill="#fff" textLength="630">Apache-2.0</text></g></svg> <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="120" height="20" role="img" aria-label="license: Apache-2.0"><title>license: Apache-2.0</title><linearGradient id="s" x2="0" y2="100%"><stop offset="0" stop-color="#bbb" stop-opacity=".1"/><stop offset="1" stop-opacity=".1"/></linearGradient><clipPath id="r"><rect width="120" height="20" rx="3" fill="#fff"/></clipPath><g clip-path="url(#r)"><rect width="47" height="20" fill="#555"/><rect x="47" width="73" height="20" fill="#97ca00"/><rect width="120" height="20" fill="url(#s)"/></g><g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" text-rendering="geometricPrecision" font-size="110"><text aria-hidden="true" x="245" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="370">license</text><text x="245" y="140" transform="scale(.1)" fill="#fff" textLength="370">license</text><text aria-hidden="true" x="825" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="630">Apache-2.0</text><text x="825" y="140" transform="scale(.1)" fill="#fff" textLength="630">Apache-2.0</text></g></svg>

Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 1.1 KiB

After

Width:  |  Height:  |  Size: 1.1 KiB

2
assets/logo.svg
View File

@ -20,4 +20,4 @@
<path d="M11.0557 19.9505L11.0406 19.9153C11.0003 19.823 10.9651 19.7291 10.9366 19.6368L10.851 19.3616L10.2856 19.3801L10.242 19.8549C10.2453 19.8733 10.2554 19.9186 10.2655 19.9992V20.016C10.2705 20.0445 10.2722 20.068 10.2772 20.0915C10.2856 20.3431 10.2101 20.5428 10.039 20.7189C9.87623 20.8917 9.66315 20.9689 9.38297 20.9572C9.03568 20.9437 8.91992 20.8045 8.86287 20.7005C8.7102 20.4253 8.70013 20.2777 8.70852 20.2122C8.73033 20.0327 8.76724 19.8431 8.81925 19.6519C8.89978 19.3532 8.83939 19.0059 8.63638 18.6217L8.42833 18.2275C8.34948 18.0782 8.28069 17.8902 8.22365 17.6688C8.18003 17.4826 8.19178 17.2594 8.25721 17.0027C8.31593 16.7762 8.44176 16.5984 8.64141 16.4574C8.84106 16.3232 9.0793 16.2746 9.36284 16.3115L9.38465 16.3148H9.40646C9.68497 16.3165 9.91314 16.3434 10.0876 16.3903C10.039 16.1185 10.0239 15.835 10.049 15.5464C9.86617 15.4693 9.68162 15.4156 9.49539 15.3904H9.48532L9.34774 15.3787C8.93334 15.3401 8.529 15.4323 8.14647 15.6538C7.75052 15.8853 7.48377 16.2276 7.35458 16.6722C7.24385 17.0598 7.23043 17.4591 7.31264 17.8567C7.3227 17.9171 7.33612 17.9876 7.35961 18.0547C7.42168 18.251 7.52403 18.4691 7.66496 18.704C7.87971 19.058 7.94515 19.1737 7.9636 19.2123L7.97031 19.2257C7.99715 19.2744 8.00219 19.3482 7.98373 19.4455L7.97702 19.4774C7.95688 19.5864 7.93676 19.6888 7.92837 19.7274C7.76059 20.1619 7.74381 20.5629 7.87636 20.9135C8.01058 21.286 8.24714 21.5578 8.58269 21.7256C8.77731 21.8212 9.00213 21.8816 9.25211 21.9068C9.38801 21.9202 9.53061 21.9219 9.67993 21.9135C10.1648 21.8883 10.5591 21.6585 10.8191 21.2541C11.0741 20.8699 11.1564 20.4438 11.0674 19.9891L11.0607 19.9505H11.0557Z" fill="#858599"/> <path d="M11.0557 19.9505L11.0406 19.9153C11.0003 19.823 10.9651 19.7291 10.9366 19.6368L10.851 19.3616L10.2856 19.3801L10.242 19.8549C10.2453 19.8733 10.2554 19.9186 10.2655 19.9992V20.016C10.2705 20.0445 10.2722 20.068 10.2772 20.0915C10.2856 20.3431 10.2101 20.5428 10.039 20.7189C9.87623 20.8917 9.66315 20.9689 9.38297 20.9572C9.03568 20.9437 8.91992 20.8045 8.86287 20.7005C8.7102 20.4253 8.70013 20.2777 8.70852 20.2122C8.73033 20.0327 8.76724 19.8431 8.81925 19.6519C8.89978 19.3532 8.83939 19.0059 8.63638 18.6217L8.42833 18.2275C8.34948 18.0782 8.28069 17.8902 8.22365 17.6688C8.18003 17.4826 8.19178 17.2594 8.25721 17.0027C8.31593 16.7762 8.44176 16.5984 8.64141 16.4574C8.84106 16.3232 9.0793 16.2746 9.36284 16.3115L9.38465 16.3148H9.40646C9.68497 16.3165 9.91314 16.3434 10.0876 16.3903C10.039 16.1185 10.0239 15.835 10.049 15.5464C9.86617 15.4693 9.68162 15.4156 9.49539 15.3904H9.48532L9.34774 15.3787C8.93334 15.3401 8.529 15.4323 8.14647 15.6538C7.75052 15.8853 7.48377 16.2276 7.35458 16.6722C7.24385 17.0598 7.23043 17.4591 7.31264 17.8567C7.3227 17.9171 7.33612 17.9876 7.35961 18.0547C7.42168 18.251 7.52403 18.4691 7.66496 18.704C7.87971 19.058 7.94515 19.1737 7.9636 19.2123L7.97031 19.2257C7.99715 19.2744 8.00219 19.3482 7.98373 19.4455L7.97702 19.4774C7.95688 19.5864 7.93676 19.6888 7.92837 19.7274C7.76059 20.1619 7.74381 20.5629 7.87636 20.9135C8.01058 21.286 8.24714 21.5578 8.58269 21.7256C8.77731 21.8212 9.00213 21.8816 9.25211 21.9068C9.38801 21.9202 9.53061 21.9219 9.67993 21.9135C10.1648 21.8883 10.5591 21.6585 10.8191 21.2541C11.0741 20.8699 11.1564 20.4438 11.0674 19.9891L11.0607 19.9505H11.0557Z" fill="#858599"/>
<path d="M31.8916 16.085C31.7607 15.6454 31.4889 15.3048 31.0829 15.0716H31.0813C30.7172 14.8636 30.328 14.778 29.9219 14.8166C29.9068 14.8166 29.8917 14.82 29.8766 14.8216L29.7424 14.8384H29.7307C29.4723 14.8804 29.2139 14.9777 28.9623 15.1236C28.9639 15.1438 28.9673 15.1656 28.9706 15.1857C28.9941 15.4424 28.9824 15.6958 28.9388 15.9424L29.0847 15.8837C29.271 15.8082 29.5243 15.7461 29.8582 15.6958L29.8766 15.6924C30.1317 15.6404 30.3514 15.6874 30.5662 15.8434C30.7927 16.0078 30.9454 16.2125 31.0326 16.4658C31.1047 16.7041 31.1148 16.9088 31.0645 17.09C31.0024 17.3165 30.9319 17.5178 30.8548 17.6872C30.7625 17.8819 30.6685 18.0714 30.5763 18.2443C30.437 18.5026 30.3867 18.7828 30.4286 19.0747C30.4638 19.3298 30.5008 19.5412 30.5427 19.7224C30.5729 19.8482 30.5544 20.0092 30.489 20.2005L30.4857 20.2089C30.432 20.38 30.1434 20.4723 29.9756 20.5126C29.5931 20.6032 29.422 20.4639 29.3196 20.3381C29.1451 20.1133 29.0629 19.87 29.068 19.5965L29.0747 19.5177C29.0831 19.4388 29.0914 19.3985 29.0948 19.3834L29.0831 18.8935L28.4556 18.9086L28.3885 19.1771C28.365 19.2694 28.3348 19.36 28.2962 19.4506L28.2794 19.4891L28.2727 19.5311C28.1989 19.9774 28.2912 20.4018 28.5445 20.7944L28.5512 20.8028C28.828 21.2021 29.224 21.4219 29.6954 21.4387C29.8213 21.4454 29.9437 21.442 30.0595 21.432C30.338 21.4068 30.5863 21.333 30.7994 21.2139C31.1131 21.0411 31.3363 20.7642 31.4671 20.3901C31.5913 20.0227 31.5661 19.6234 31.3933 19.2023L31.3514 18.9539L31.348 18.9355C31.3279 18.8466 31.3363 18.7593 31.3698 18.6721C31.4353 18.516 31.5242 18.3349 31.6332 18.1285C31.7758 17.8768 31.8681 17.6638 31.9184 17.4826C31.9436 17.402 31.9604 17.3316 31.9654 17.2611C32.0309 16.8668 32.0057 16.4726 31.8882 16.0884L31.8916 16.085Z" fill="#858599"/> <path d="M31.8916 16.085C31.7607 15.6454 31.4889 15.3048 31.0829 15.0716H31.0813C30.7172 14.8636 30.328 14.778 29.9219 14.8166C29.9068 14.8166 29.8917 14.82 29.8766 14.8216L29.7424 14.8384H29.7307C29.4723 14.8804 29.2139 14.9777 28.9623 15.1236C28.9639 15.1438 28.9673 15.1656 28.9706 15.1857C28.9941 15.4424 28.9824 15.6958 28.9388 15.9424L29.0847 15.8837C29.271 15.8082 29.5243 15.7461 29.8582 15.6958L29.8766 15.6924C30.1317 15.6404 30.3514 15.6874 30.5662 15.8434C30.7927 16.0078 30.9454 16.2125 31.0326 16.4658C31.1047 16.7041 31.1148 16.9088 31.0645 17.09C31.0024 17.3165 30.9319 17.5178 30.8548 17.6872C30.7625 17.8819 30.6685 18.0714 30.5763 18.2443C30.437 18.5026 30.3867 18.7828 30.4286 19.0747C30.4638 19.3298 30.5008 19.5412 30.5427 19.7224C30.5729 19.8482 30.5544 20.0092 30.489 20.2005L30.4857 20.2089C30.432 20.38 30.1434 20.4723 29.9756 20.5126C29.5931 20.6032 29.422 20.4639 29.3196 20.3381C29.1451 20.1133 29.0629 19.87 29.068 19.5965L29.0747 19.5177C29.0831 19.4388 29.0914 19.3985 29.0948 19.3834L29.0831 18.8935L28.4556 18.9086L28.3885 19.1771C28.365 19.2694 28.3348 19.36 28.2962 19.4506L28.2794 19.4891L28.2727 19.5311C28.1989 19.9774 28.2912 20.4018 28.5445 20.7944L28.5512 20.8028C28.828 21.2021 29.224 21.4219 29.6954 21.4387C29.8213 21.4454 29.9437 21.442 30.0595 21.432C30.338 21.4068 30.5863 21.333 30.7994 21.2139C31.1131 21.0411 31.3363 20.7642 31.4671 20.3901C31.5913 20.0227 31.5661 19.6234 31.3933 19.2023L31.3514 18.9539L31.348 18.9355C31.3279 18.8466 31.3363 18.7593 31.3698 18.6721C31.4353 18.516 31.5242 18.3349 31.6332 18.1285C31.7758 17.8768 31.8681 17.6638 31.9184 17.4826C31.9436 17.402 31.9604 17.3316 31.9654 17.2611C32.0309 16.8668 32.0057 16.4726 31.8882 16.0884L31.8916 16.085Z" fill="#858599"/>
<path d="M24.1318 4.14939C24.1804 3.88095 24.2089 3.62761 24.219 3.37427C24.2358 2.43137 23.7811 1.61599 22.8651 0.94992C21.9809 0.329153 20.9122 0.010386 19.6891 0.000319559C18.4627 -0.0114247 17.399 0.300635 16.5249 0.921401C15.6105 1.57404 15.1407 2.37768 15.129 3.31051V3.3189C15.1307 3.50513 15.1424 3.69136 15.1659 3.88598C15.0803 3.98832 15.0334 4.10073 15.0283 4.22153V4.23495C15.0283 4.55037 15.0736 4.86578 15.1692 5.17113L15.1709 5.17952C15.3018 5.56541 15.5266 5.93283 15.8403 6.27173L15.9309 6.36905L16.7564 6.46132C17.461 6.87069 18.4023 7.09048 19.5599 7.11061C19.6102 7.11061 19.6606 7.11061 19.7092 7.11061C20.5632 7.11061 21.3098 7.00659 21.9306 6.8019L21.9473 6.7952C22.1805 6.7046 22.407 6.59554 22.6235 6.46971C23.2812 6.53179 23.7744 6.20798 24.0277 5.54863L24.0898 5.48152L24.0865 5.38756C24.1838 5.08892 24.3549 4.50339 24.1301 4.14939H24.1318ZM16.9359 1.50022C17.6842 0.968375 18.6069 0.703294 19.6824 0.711683C20.7578 0.720071 21.6906 0.996899 22.4507 1.52874C23.1738 2.05388 23.5194 2.65451 23.5076 3.35413C23.5026 3.48332 23.4925 3.61586 23.4741 3.75344C22.2527 3.24173 20.9491 2.98671 19.5683 2.98671C18.3821 2.98671 17.1406 3.17629 15.8487 3.55379C15.8437 3.47326 15.8404 3.39272 15.8387 3.31387C15.8487 2.61257 16.2078 2.01864 16.9359 1.50022ZM23.5009 4.83558L23.261 5.35401C23.261 5.35401 22.9473 5.807 22.605 5.74996L22.4725 5.72815L22.3584 5.79861C22.1504 5.9278 21.9289 6.03853 21.699 6.12745C21.1202 6.31704 20.4038 6.40764 19.5717 6.39589C18.5415 6.37576 17.7161 6.1912 17.1171 5.84391C17.0987 5.83217 17.0802 5.82042 17.0634 5.81203L16.993 5.76841L16.2782 5.68956C16.1423 5.53017 15.9494 5.19462 15.9494 5.19462L15.7531 4.62586C15.7531 4.62586 15.743 4.38427 15.7414 4.32891C18.6338 3.41789 21.2544 3.49339 23.5311 4.55372C23.5345 4.59567 23.5009 4.83558 23.5009 4.83558ZM23.714 5.36407V5.35904L23.719 5.36407H23.714Z" fill="#858599"/> <path d="M24.1318 4.14939C24.1804 3.88095 24.2089 3.62761 24.219 3.37427C24.2358 2.43137 23.7811 1.61599 22.8651 0.94992C21.9809 0.329153 20.9122 0.010386 19.6891 0.000319559C18.4627 -0.0114247 17.399 0.300635 16.5249 0.921401C15.6105 1.57404 15.1407 2.37768 15.129 3.31051V3.3189C15.1307 3.50513 15.1424 3.69136 15.1659 3.88598C15.0803 3.98832 15.0334 4.10073 15.0283 4.22153V4.23495C15.0283 4.55037 15.0736 4.86578 15.1692 5.17113L15.1709 5.17952C15.3018 5.56541 15.5266 5.93283 15.8403 6.27173L15.9309 6.36905L16.7564 6.46132C17.461 6.87069 18.4023 7.09048 19.5599 7.11061C19.6102 7.11061 19.6606 7.11061 19.7092 7.11061C20.5632 7.11061 21.3098 7.00659 21.9306 6.8019L21.9473 6.7952C22.1805 6.7046 22.407 6.59554 22.6235 6.46971C23.2812 6.53179 23.7744 6.20798 24.0277 5.54863L24.0898 5.48152L24.0865 5.38756C24.1838 5.08892 24.3549 4.50339 24.1301 4.14939H24.1318ZM16.9359 1.50022C17.6842 0.968375 18.6069 0.703294 19.6824 0.711683C20.7578 0.720071 21.6906 0.996899 22.4507 1.52874C23.1738 2.05388 23.5194 2.65451 23.5076 3.35413C23.5026 3.48332 23.4925 3.61586 23.4741 3.75344C22.2527 3.24173 20.9491 2.98671 19.5683 2.98671C18.3821 2.98671 17.1406 3.17629 15.8487 3.55379C15.8437 3.47326 15.8404 3.39272 15.8387 3.31387C15.8487 2.61257 16.2078 2.01864 16.9359 1.50022ZM23.5009 4.83558L23.261 5.35401C23.261 5.35401 22.9473 5.807 22.605 5.74996L22.4725 5.72815L22.3584 5.79861C22.1504 5.9278 21.9289 6.03853 21.699 6.12745C21.1202 6.31704 20.4038 6.40764 19.5717 6.39589C18.5415 6.37576 17.7161 6.1912 17.1171 5.84391C17.0987 5.83217 17.0802 5.82042 17.0634 5.81203L16.993 5.76841L16.2782 5.68956C16.1423 5.53017 15.9494 5.19462 15.9494 5.19462L15.7531 4.62586C15.7531 4.62586 15.743 4.38427 15.7414 4.32891C18.6338 3.41789 21.2544 3.49339 23.5311 4.55372C23.5345 4.59567 23.5009 4.83558 23.5009 4.83558ZM23.714 5.36407V5.35904L23.719 5.36407H23.714Z" fill="#858599"/>
</svg> </svg>

Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 18 KiB

After

Width:  |  Height:  |  Size: 18 KiB

6
chat/openaoe.md
View File

@ -12,10 +12,10 @@ Currently already supported LLMs: [InternLM2-Chat-7B](https://huggingface.co/int
We provide three different ways to run OpenAOE: `run by pip`， `run by docker` and `run by source code` as well. We provide three different ways to run OpenAOE: `run by pip`， `run by docker` and `run by source code` as well.
### Run by pip ### Run by pip
#### **Install** #### **Install**
```shell ```shell
pip install -U openaoe pip install -U openaoe
``` ```
#### **Start** #### **Start**
```shell ```shell
@ -65,7 +65,7 @@ python -m main -f /path/to/your/config-template.yaml
``` ```
> [!TIP] > [!TIP]
> `/path/to/your/config.yaml` is the configuration file loaded by OpenAOE at startup, > `/path/to/your/config.yaml` is the configuration file loaded by OpenAOE at startup,
> which contains the relevant configuration information for the LLMs, > which contains the relevant configuration information for the LLMs,
> including: API URLs, AKSKs, Tokens, etc. > including: API URLs, AKSKs, Tokens, etc.
> A template configuration yaml file can be found in `openaoe/backend/config/config.yaml`. > A template configuration yaml file can be found in `openaoe/backend/config/config.yaml`.

4
chat/openaoe_zh_cn.md
View File

@ -17,7 +17,7 @@
> 需要 python >= 3.9 > 需要 python >= 3.9
#### **安装** #### **安装**
```shell ```shell
pip install -U openaoe pip install -U openaoe
``` ```
#### **运行** #### **运行**
```shell ```shell
@ -50,7 +50,7 @@ docker run -p 10099:10099 -v /path/to/your/config-template.yaml:/app/config-temp
```shell ```shell
git clone https://github.com/internlm/OpenAOE git clone https://github.com/internlm/OpenAOE
``` ```
2. [_可选_] （如果前端代码发生变动）重新构建前端项目 2. [_可选_] （如果前端代码发生变动）重新构建前端项目
```shell ```shell
cd open-aoe/openaoe/frontend cd open-aoe/openaoe/frontend
npm install npm install

13
chat/web_demo.py
View File

@ -11,11 +11,10 @@ from dataclasses import asdict
import streamlit as st import streamlit as st
import torch import torch
from tools.transformers.interface import GenerationConfig, generate_interactive
from transformers import AutoModelForCausalLM, AutoTokenizer from transformers import AutoModelForCausalLM, AutoTokenizer
from transformers.utils import logging from transformers.utils import logging
from tools.transformers.interface import GenerationConfig, generate_interactive
logger = logging.get_logger(__name__) logger = logging.get_logger(__name__)
@ -109,9 +108,15 @@ def main():
): ):
# Display robot response in chat message container # Display robot response in chat message container
message_placeholder.markdown(cur_response + "▌") message_placeholder.markdown(cur_response + "▌")
message_placeholder.markdown(cur_response) message_placeholder.markdown(cur_response) # pylint: disable=undefined-loop-variable
# Add robot response to chat history # Add robot response to chat history
st.session_state.messages.append({"role": "robot", "content": cur_response, "avatar": robot_avator}) st.session_state.messages.append(
{
"role": "robot",
"content": cur_response, # pylint: disable=undefined-loop-variable
"avatar": robot_avator,
}
)
torch.cuda.empty_cache() torch.cuda.empty_cache()

2
model_cards/internlm2_20b.md
View File

@ -38,5 +38,5 @@ We have evaluated InternLM2 on several important benchmarks using the open-sourc
| MBPP(Sanitized) | 51.8 | 51.4 | 63.0 | 65.8 | 78.9 | 79.0 | | MBPP(Sanitized) | 51.8 | 51.4 | 63.0 | 65.8 | 78.9 | 79.0 |
- The evaluation results were obtained from [OpenCompass](https://github.com/open-compass/opencompass) , and evaluation configuration can be found in the configuration files provided by [OpenCompass](https://github.com/open-compass/opencompass). - The evaluation results were obtained from [OpenCompass](https://github.com/open-compass/opencompass) , and evaluation configuration can be found in the configuration files provided by [OpenCompass](https://github.com/open-compass/opencompass).
- The evaluation data may have numerical differences due to the version iteration of [OpenCompass](https://github.com/open-compass/opencompass), so please refer to the latest evaluation results of [OpenCompass](https://github.com/open-compass/opencompass). - The evaluation data may have numerical differences due to the version iteration of [OpenCompass](https://github.com/open-compass/opencompass), so please refer to the latest evaluation results of [OpenCompass](https://github.com/open-compass/opencompass).

2
model_cards/internlm2_7b.md
View File

@ -38,5 +38,5 @@ We have evaluated InternLM2 on several important benchmarks using the open-sourc
| MBPP(Sanitized) | 51.8 | 51.4 | 63.0 | 65.8 | 78.9 | 79.0 | | MBPP(Sanitized) | 51.8 | 51.4 | 63.0 | 65.8 | 78.9 | 79.0 |
- The evaluation results were obtained from [OpenCompass](https://github.com/open-compass/opencompass) , and evaluation configuration can be found in the configuration files provided by [OpenCompass](https://github.com/open-compass/opencompass). - The evaluation results were obtained from [OpenCompass](https://github.com/open-compass/opencompass) , and evaluation configuration can be found in the configuration files provided by [OpenCompass](https://github.com/open-compass/opencompass).
- The evaluation data may have numerical differences due to the version iteration of [OpenCompass](https://github.com/open-compass/opencompass), so please refer to the latest evaluation results of [OpenCompass](https://github.com/open-compass/opencompass). - The evaluation data may have numerical differences due to the version iteration of [OpenCompass](https://github.com/open-compass/opencompass), so please refer to the latest evaluation results of [OpenCompass](https://github.com/open-compass/opencompass).