salesforce/jaxformer issues and pull requests

#32 - Bump transformers from 4.9.2 to 4.36.0 in /preprocess

Pull Request - State: open - Opened by dependabot[bot] 11 months ago
Labels: dependencies

#31 - facing issue while running jaxformer locally.

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

#30 - Bump transformers from 4.9.2 to 4.30.0 in /preprocess

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: dependencies

#29 - added deepspeed inference

Pull Request - State: closed - Opened by mlap1n over 1 year ago - 1 comment
Labels: cla:missing

#28 - Weight decay is applied to the layernorm in transformer block

Issue - State: open - Opened by sh0416 over 1 year ago

#26 - Log Z term in loss

Issue - State: open - Opened by sh0416 over 1 year ago - 2 comments

#25 - Pipeline Parallelism

Issue - State: closed - Opened by sh0416 over 1 year ago - 1 comment

#24 - Request for training configuration of CodeGen 16B

Issue - State: open - Opened by sh0416 over 1 year ago - 1 comment

#23 - Memory to fine-tune 16B model?

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

#22 - Bump tensorflow from 2.7.2 to 2.11.1 in /preprocess

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#21 - Bump tensorflow-cpu from 2.7.2 to 2.11.1

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#20 - [Mismatched_sizes] Got mismatched_size exception when loading the finetuned model

Issue - State: open - Opened by Jacob-yen over 1 year ago

#19 - TPU finetuned model corrupts

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

#18 - Is the BigQuery dataset public available?

Issue - State: open - Opened by HongtaoYang almost 2 years ago - 4 comments

#17 - how to make tfrecord with a released tokenizer

Issue - State: open - Opened by HaebinShin almost 2 years ago

#16 - can't find paper

Issue - State: open - Opened by glicerico almost 2 years ago - 1 comment

#15 - Fine-tuning on conversations (format of conversations)

Issue - State: open - Opened by Eichhof almost 2 years ago - 1 comment

#14 - [Suggestion]: Code Notes

Issue - State: open - Opened by Librechain almost 2 years ago - 2 comments

#13 - How to load codegen models in jaxformer locally??

Issue - State: open - Opened by harry-stark about 2 years ago - 1 comment

#12 - Bump tensorflow from 2.7.2 to 2.9.3 in /preprocess

Pull Request - State: closed - Opened by dependabot[bot] about 2 years ago - 1 comment
Labels: dependencies

#11 - Bump tensorflow-cpu from 2.7.2 to 2.9.3

Pull Request - State: closed - Opened by dependabot[bot] about 2 years ago - 1 comment
Labels: dependencies

#10 - Out-of-memory running with Deepspeed

Issue - State: open - Opened by calix about 2 years ago - 2 comments

#9 - Can we use Jaxformer for Nvidia A100 GPU currently?

Issue - State: open - Opened by leemgs about 2 years ago - 3 comments

#8 - Add <endoftext> for every lines

Issue - State: open - Opened by PoodleWang about 2 years ago - 5 comments

#7 - Cannot find 16B jax checkpoint

Issue - State: open - Opened by PoodleWang about 2 years ago - 2 comments

#6 - 350M Mono not found

Issue - State: open - Opened by bycn about 2 years ago - 3 comments

#5 - ISSUE: thread source out

Issue - State: closed - Opened by PoodleWang about 2 years ago - 1 comment

#4 - issue: why you comment this line? Do you still need the bos?