Skip to content
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.

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

Adapted megatronlm server implementation for interacting with lm eval… #8

Open
wants to merge 30 commits into
base: main
Choose a base branch
from

Conversation

KlaudiaTH
Copy link

… harness client

@@ -43,7 +43,8 @@ def build_tokenizer(args):
elif args.tokenizer_type == "OpenGPTX-PretrainedHFTokenizer":
tokenizer = PretrainedHFTokenizer.instantiate_from_file_or_name(model_file_or_name=args.tokenizer_model)
elif args.tokenizer_type == "OpenGPTX-SPTokenizer":
tokenizer = SPTokenizer.instantiate_from_file_or_name(model_file_or_name=args.tokenizer_model)
#tokenizer = SPTokenizer.instantiate_from_file_or_name(model_file_or_name=args.tokenizer_model)
tokenizer = _GPTXSentencePieceTokenizer(args.tokenizer_model)
Copy link
Author

@KlaudiaTH KlaudiaTH Aug 3, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ToDo: The _GPTXSentencePieceTokenizer is used for inference.

@@ -164,48 +179,60 @@ def put(self):
if len(prompts) > 1:
return "When doing beam_search, batch size must be 1"

stop_token=50256
stop_token = 3
Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Setting this to a constant 3 is wrong. I haven't tested generation yet, but I guess this should be the ID of the EOD token, so we need to get it from the tokenizer.

Copy link

github-actions bot commented Jan 5, 2024

Marking as stale. No activity in 60 days.

@github-actions github-actions bot added the stale label Jan 5, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants