Skip to content

chujiezheng/chat_templates

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

82 Commits
chat_templates
chat_templates
 
 
generation_configs
generation_configs
 
 
LICENSE
LICENSE
 
 
README.md
README.md
 
 

Repository files navigation

chat_templates

This is a repository that includes proper chat templates (or input formats) for instruction-tuned large language models (LLMs), to support transformers's chat_template feature. If you are interested to include more chat templates, feel free to open a pull request

If you find this repo useful, please kindly cite it:

@misc{zheng-2024-chat-templates,
  author = {Zheng, Chujie},
  title = {Chat Templates for HuggingFace Large Language Models},
  year = {2024},
  howpublished = {\url{https://github.com/chujiezheng/chat_templates}}
}

Updates

  • [10/2024] Added support for IBM's Granite-3.0 models
  • [07/2024] Added support for Meta's Llama-3.1 models
  • [06/2024] Added support for Google's Gemma-2 models
  • [05/2024] Added support for Nvidia's ChatQA models
  • [04/2024] Added support for Microsoft's Phi-3 models
  • [04/2024] Added support for Meta's Llama-3 models
  • [02/2024] Added support for Google's Gemma models
  • [02/2024] Added usage explanation for generation_configs
  • [01/2024] Added support for Alibaba's Qwen2 models

What are Contained in This Repo?

  • chat_templates contains the jinja files of collected chat templates, which can be directly replaced in the Huggingface tokenizers

  • generation_configs contains the corresponding json configs used for controlling the ending of response generations. Specially, the stop_token_ids should be directly passed into the generate method by the eos_token_id argument

Usage Examples

Important NOTE: As mentioned in this issue, the messages should contain at least one user message. It is strongly not recommented to pass only the system message, as there may result in unexpected outputs (because the models are not trained in this way).

Example 1: Meta-Llama-3-8B-Instruct

This example may check if the jinja file is correctly implemented.

from transformers import AutoTokenizer

toker = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B-Instruct", token="YOUR_OWN_TOKEN")
messages = [
    {'role': 'system', 'content': 'This is a system prompt.'},
    {'role': 'user', 'content': 'This is the first user input.'},
    {'role': 'assistant', 'content': 'This is the first assistant response.'},
    {'role': 'user', 'content': 'This is the second user input.'},
]
print('###### Default (yet Correct) Chat Template ######')
print(toker.apply_chat_template(messages, tokenize=False, add_generation_prompt=True))
print('###### Corrected Chat Template ######')
chat_template = open('./chat_templates/llama-3-instruct.jinja').read()
chat_template = chat_template.replace('    ', '').replace('\n', '')
toker.chat_template = chat_template
print(toker.apply_chat_template(messages, tokenize=False, add_generation_prompt=True))

Expected output:

###### Default (yet Correct) Chat Template ######
<|begin_of_text|><|start_header_id|>system<|end_header_id|>

This is a system prompt.<|eot_id|><|start_header_id|>user<|end_header_id|>

This is the first user input.<|eot_id|><|start_header_id|>assistant<|end_header_id|>

This is the first assistant response.<|eot_id|><|start_header_id|>user<|end_header_id|>

This is the second user input.<|eot_id|><|start_header_id|>assistant<|end_header_id|>


###### Corrected Chat Template ######
<|begin_of_text|><|start_header_id|>system<|end_header_id|>

This is a system prompt.<|eot_id|><|start_header_id|>user<|end_header_id|>

This is the first user input.<|eot_id|><|start_header_id|>assistant<|end_header_id|>

This is the first assistant response.<|eot_id|><|start_header_id|>user<|end_header_id|>

This is the second user input.<|eot_id|><|start_header_id|>assistant<|end_header_id|>
Example 2: Mistral-7B-Instruct-v0.2

For mistral-instruct (also gemma-it), it does not natively support the system message, so passing the system message would raise error.

from transformers import AutoTokenizer

toker = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
messages = [
    {'role': 'system', 'content': 'This is a system prompt.'},
    {'role': 'user', 'content': 'This is the first user input.'},
    {'role': 'assistant', 'content': 'This is the first assistant response.'},
    {'role': 'user', 'content': 'This is the second user input.'},
]
print('###### Default (but Improper) Chat Template ######')
# raising error
#print(toker.apply_chat_template(messages, tokenize=False, add_generation_prompt=True))
print('###### Corrected Chat Template ######')
chat_template = open('./chat_templates/mistral-instruct.jinja').read()
chat_template = chat_template.replace('    ', '').replace('\n', '')
toker.chat_template = chat_template
print(toker.apply_chat_template(messages, tokenize=False, add_generation_prompt=True))

Expected output:

###### Default (but Error-Raising) Chat Template ######
jinja2.exceptions.TemplateError: Conversation roles must alternate user/assistant/user/assistant/...
###### Corrected Chat Template ######
<s>[INST] This is a system prompt.

This is the first user input. [/INST] This is the first assistant response. </s>[INST] This is the second user input. [/INST]
Example 3: vicuna-7b-v1.5

NOTE: In fast-chat, vicuna does not add linebreaks between roles' messages. But I found that adding linebreaks leads to a bit better performance (especially for the v1.5 version).

Also, I found vicuna-7/13/33b-v1.3 may not work well when given a system message different from its default one. So I would recommend to use vicuna-7/13b-v1.5 instead.

from transformers import AutoTokenizer

toker = AutoTokenizer.from_pretrained("lmsys/vicuna-7b-v1.5")
messages = [
    {'role': 'system', 'content': 'This is a system prompt.'},
    {'role': 'user', 'content': 'This is the first user input.'},
    {'role': 'assistant', 'content': 'This is the first assistant response.'},
    {'role': 'user', 'content': 'This is the second user input.'},
]
print('###### Default (but Improper) Chat Template ######')
print(toker.apply_chat_template(messages, tokenize=False, add_generation_prompt=True))
print('###### Corrected Chat Template ######')
chat_template = open('./chat_templates/vicuna.jinja').read()
chat_template = chat_template.replace('    ', '').replace('\n', '')
toker.chat_template = chat_template
print(toker.apply_chat_template(messages, tokenize=False, add_generation_prompt=True))

Expected output:

###### Default (but Improper) Chat Template ######
<s>[INST] <<SYS>>
This is a system prompt.
<</SYS>>

This is the first user input. [/INST] This is the first assistant response. </s><s>[INST] This is the second user input. [/INST]
###### Corrected Chat Template ######
<s>This is a system prompt.

USER: This is the first user input.
ASSISTANT: This is the first assistant response.</s>
USER: This is the second user input.
ASSISTANT:

Supported Models

NOTE: The listed models are not inclusive and also include other-sized ones in the same model family

granite-3.0-instruct
Llama-3.1-Instruct
Llama-3-Instruct
Llama-2-Chat, CodeLlama-Instruct
Qwen2-Instruct, Qwen1.5-Chat
Mistral-Instruct
Phi-3-Instruct
Yi-1.5-Chat, Yi-Chat
gemma-it, gemma-2-it
Llama3-ChatQA-1.5
openchat-3.5, Starling-LM
zephyr
vicuna
Orca-2
falcon-instruct
SOLAR-Instruct
Alpaca
AmberChat
saiga

Star History

Star History Chart

About

Chat Templates for 🤗 HuggingFace Large Language Models

Resources

Readme

License

MIT license
Activity

Stars

538 stars

Watchers

9 watching

Forks

53 forks
Report repository

Releases

No releases published

Packages

No packages published

Contributors 11

Languages