Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / epfml/landmark-attention issues and pull requests

#16 - Fine-tuned Weights Doesn't Work

Issue - State: open - Opened by baotruyenthach 8 months ago

#14 - Question about training stability

Issue - State: open - Opened by meme-virus over 1 year ago - 3 comments

#12 - when using auto_insert_landmarks, it appears to be view problems

Issue - State: open - Opened by EzTan1 over 1 year ago - 1 comment

#11 - Assertion `srcIndex < srcSelectDimSize` failed while running test.

Issue - State: closed - Opened by L16H7 over 1 year ago - 2 comments

#10 - Peft module

Issue - State: closed - Opened by NicolasMejiaPetit over 1 year ago

#9 - How much VRAM do you need to run Inference?

Issue - State: open - Opened by FFFiend over 1 year ago

#8 - Unlimited context length and no prompt token limit

Issue - State: closed - Opened by ekolawole over 1 year ago

#7 - Update README.md

Pull Request - State: open - Opened by eltociear over 1 year ago

#6 - Question about Training vs LLaMA Fine Tuning

Issue - State: closed - Opened by eugenepentland over 1 year ago - 2 comments

#4 - model_name_or_path defaults

Issue - State: open - Opened by StrangeTcy over 1 year ago

#3 - Readme clarification

Issue - State: closed - Opened by StrangeTcy over 1 year ago - 1 comment

#2 - Is true attention (full-context) calculation needed during training?

Issue - State: closed - Opened by Andrey36652 over 1 year ago - 3 comments

#1 - Will this work with Qlora and 4bit inference?

Issue - State: open - Opened by jordancole21 over 1 year ago - 36 comments