Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Adds more stop-tokens to chat formatter #699

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
JamesKunstle wants to merge 1 commit into from
Closed

Adds more stop-tokens to chat formatter #699

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions cli/lab.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -611,6 +611,7 @@ def download(ctx, repository, release, filename, model_dir):
filename=filename,
local_dir=model_dir,
)
click.echo("==> Model ready!")
except Exception as exc:
click.secho(
f"Downloading model failed with the following Hugging Face Hub error: {exc}",
Expand Down
17 changes: 15 additions & 2 deletions cli/server.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -127,11 +127,24 @@ def server(
raise ServerException(f"failed creating the server application: {exc}") from exc

try:
llama_app._llama_proxy._current_model.chat_handler = llama_chat_format.Jinja2ChatFormatter(
chatFormatterResponse = llama_chat_format.Jinja2ChatFormatter(
template="{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
eos_token="<|endoftext|>",
bos_token="",
).to_chat_handler()
)

# add extra stop tokens because Jinja2ChatFormatter only accepts one for template.
chatFormatterResponse.stop.extend(
[
"<|system|>",
"<|user|>",
"<|assistant|>",
]
)

llama_app._llama_proxy._current_model.chat_handler = (
chatFormatterResponse.to_chat_handler()
)
# pylint: disable=broad-exception-caught
except Exception as exc:
if queue:
Expand Down