emrgnt-cmplxty/smoltrainer issues and pull requests

#38 - relocate scripts and add some new one

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

#38 - relocate scripts and add some new one

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

#37 - Extend data creation pipe

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#37 - Extend data creation pipe

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#36 - add ability to remix data quickly

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#36 - add ability to remix data quickly

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#35 - finalize

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#35 - finalize

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#34 - Fix checkpoint logic

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#34 - Fix checkpoint logic

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#33 - Feature/modify vocab

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#33 - Feature/modify vocab

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#32 - commit changes

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#32 - commit changes

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#31 - Remove tensorboard

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#31 - Remove tensorboard

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#30 - Feature/add flash v2

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#30 - Feature/add flash v2

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#29 - Feature/iterate until fix

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#29 - Feature/iterate until fix

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#28 - proper fix

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#28 - proper fix

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#27 - Feature/proper fix

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#27 - Feature/proper fix

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#26 - fix device type error

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#26 - fix device type error

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#25 - fix

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#25 - fix

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#24 - Polish

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#24 - Polish

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#23 - Feature/checkpt uploader rebased

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#23 - Feature/checkpt uploader rebased

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#22 - Feature/open orca ft dset

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#22 - Feature/open orca ft dset

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#21 - Feature/fix nans add prep

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#21 - Feature/fix nans add prep

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#20 - fix bugs, add lora and llama

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#20 - fix bugs, add lora and llama

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#19 - Feature/add lora

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#19 - Feature/add lora

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#18 - Add integrated training time to performance logging

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#18 - Add integrated training time to performance logging

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#17 - Cleanup trainer, fix wandb

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#17 - Cleanup trainer, fix wandb

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#16 - Feature/add inference

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#16 - Feature/add inference

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#15 - Feature/add tensorboard rebased

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#15 - Feature/add tensorboard rebased

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#14 - Add tensorboard

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#13 - Add checkpoint logic

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#12 - Feature/add trainer module rebased

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#11 - Feature/add trainer module

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#10 - Feature/optimize moe torch impl rebased

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#9 - Cleanup torch impl of MoE for greater eff.

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#8 - Feature/optimize moe torch impl

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#7 - Feature/add config

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#6 - Feature/attempt top k moe

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#5 - Adding in a cli, other cleanups to train.py

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#4 - Feature/add license

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#3 - Add nano_gpt and MoE model

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#2 - Add nano_gpt and MoE model

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

#1 - Add nano_gpt and MoE model

Pull Request - State: closed - Opened by emrgnt-cmplxty over 1 year ago

GitHub / emrgnt-cmplxty/smoltrainer issues and pull requests