Internlm2-chat-7b无法生成正确的回复

[AlanLu0808]

提交前必须检查以下项目 | The following items must be checked before submission

• [X] 请确保使用的是仓库最新代码（git pull），一些问题已被解决和修复。 | Make sure you are using the latest code from the repository (git pull), some issues have already been addressed and fixed.
• [X] 我已阅读项目文档和FAQ章节并且已在Issue中对问题进行了搜索，没有找到相似问题和解决方案 | I have searched the existing issues / discussions

问题类型 | Type of problem

效果问题 | Effectiveness issues

操作系统 | Operating system

Linux

详细描述问题 | Detailed description of the problem

Internlm2-chat-7b无法生成正确的回复，使用的是2024.02.04版本

配置1

PORT=8945

# model related
MODEL_NAME=internlm2
MODEL_PATH=/mnt/LLM/public/LLM/internlm2-chat-7b
EMBEDDING_NAME=
ADAPTER_MODEL_PATH=
QUANTIZE=16
CONTEXT_LEN=
LOAD_IN_8BIT=false
LOAD_IN_4BIT=false
USING_PTUNING_V2=false
STREAM_INTERVERL=2
PROMPT_NAME= 

# device related
DEVICE=

# "auto", "cuda:0", "cuda:1", ...
DEVICE_MAP=auto
GPUS=
NUM_GPUs=1
DTYPE=half

# api related
API_PREFIX=/v1

USE_STREAMER_V2=false
ENGINE=default

配置2：增加PROMPT_NAME字段

PORT=8945

# model related
MODEL_NAME=internlm2
MODEL_PATH=/mnt/LLM/public/LLM/internlm2-chat-7b
EMBEDDING_NAME=
ADAPTER_MODEL_PATH=
QUANTIZE=16
CONTEXT_LEN=
LOAD_IN_8BIT=false
LOAD_IN_4BIT=false
USING_PTUNING_V2=false
STREAM_INTERVERL=2
PROMPT_NAME=internlm2

# device related
DEVICE=

# "auto", "cuda:0", "cuda:1", ...
DEVICE_MAP=auto
GPUS=
NUM_GPUs=1
DTYPE=half

# api related
API_PREFIX=/v1

USE_STREAMER_V2=false
ENGINE=default

Dependencies

# 请在此处粘贴依赖情况
# Please paste the dependencies here

运行日志或截图 | Runtime logs or screenshots

配置1

配置2

[xusenlinzy]

https://github.com/xusenlinzy/api-for-open-llm/blob/d2a9e6c95106288756f26d6cae3f14a77ac20025/api/adapter/template.py#L775

这一行改一下：

"strings": ["</s>", "<|im_end|>"], 
"token_ids": [92541]  # <|action_start|>

试试有没有用

[AlanLu0808]

修改后InternLM2-chat-20B是正常的，但是InternLM2-chat-7B看起来效果还是不对

InternLM2-chat-7B效果如下：