Ecosyste.ms: Issues

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

GitHub / git-cloner/llama-lora-fine-tuning issues and pull requests

#16 - Replace removed function prepare_model_for_int8_training

Pull Request - State: closed - Opened by yana9i 9 months ago

#15 - About the result of lora fine-tuning

Issue - State: open - Opened by lucasliunju about 1 year ago - 2 comments

#14 - ModuleNotFoundError: No module named 'utils.prompter'

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

#13 - Run deepspeed fastchat/train/train_lora.py error. Padding Error

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

#12 - Expected a cuda device, but got: cpu

Issue - State: open - Opened by mvuthegoat over 1 year ago - 4 comments

#11 - RuntimeError:CUDA error : out of memory

Issue - State: open - Opened by Hzzhang-nlp over 1 year ago - 4 comments

#10 - Do the fine tuning and seting --model_max_length 2048 issue

Issue - State: open - Opened by JustinZou1 over 1 year ago - 1 comment

#8 - training stuck

Issue - State: open - Opened by mz2sj over 1 year ago - 1 comment

#7 - down llama model so slowly

Issue - State: closed - Opened by mz2sj over 1 year ago - 2 comments

#6 - error: subprocess-exited-with-error

Issue - State: open - Opened by Hzzhang-nlp over 1 year ago - 5 comments

#5 - How to resume correct

Issue - State: closed - Opened by ycat3 over 1 year ago - 1 comment

#4 - run deepspeed fastchat/train/train_lora.py error

Issue - State: closed - Opened by codezealot over 1 year ago - 5 comments

#3 - sequence length is longer than the specified maximum sequence length

Issue - State: closed - Opened by ycat3 over 1 year ago - 1 comment

#2 - 3.3.3 Html to Markdown never finish.

Issue - State: closed - Opened by ycat3 over 1 year ago - 2 comments

#1 - do support multi-gpus?

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