Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 12, 2023

• Minor readme updates to fix linkrot and make it clear which links download files. (#247)

  StellaAthena

  authored
  f8315e3

Commits on Jan 28, 2022

• fix #180

  kingoflolz

  committed
  0a75ca9

Commits on Jan 19, 2022

• Fix the download link for the weights (#183)

  versae

  authored
  7b3507c

Commits on Jan 17, 2022

• update (#182)

  leogao2

  authored
  04f76ba

Commits on Jan 14, 2022

• fix tqdm version conflict (#179)

  widiba03304

  authored
  a982057

Commits on Dec 12, 2021

• add Gopher 230B results (#168)

  djoldman

  authored
  e84de89

Commits on Dec 4, 2021

• fix off by one

  kingoflolz

  committed
  d2c2f59

• fix deps

  kingoflolz

  committed
  628f44d

Commits on Nov 13, 2021

• Update requirements.txt (#151)

  reouno

  authored
  fa5abc5

Commits on Nov 11, 2021

• Add Megatron-Turing NLG 530B numbers (#138)

  djoldman

  authored
  56f8b74

• bump tensorflow dep

  kingoflolz

  committed
  27a8036

Commits on Oct 18, 2021

• added support for tfrecord from single file. (#140)

  ablacklama

  authored
  1ac79c1

Commits on Sep 24, 2021

• Fixing config saving bug in to_hf_weights.py + adding pathy to requirements (#131)

  ablacklama

  authored
  592f6ff

Commits on Sep 15, 2021

• Lower memory consumption in Colab demo (#126)

  vfbd

  authored
  8f071d0

Commits on Sep 11, 2021

• update readme (#121)

  rozanecm

  authored
  0c32f0c

• clarify val_batches (#122)

  nostalgebraist

  authored
  794c1b6

Commits on Sep 1, 2021

• Update howto_finetune.md (#116)

  StellaAthena

  authored
  e402470

Commits on Aug 27, 2021

• attempt inference with v2 model

  kingoflolz

  committed
  0f9b555

Commits on Aug 25, 2021

• Dockerfile requires apt update before git install (#111)

  jacobfnl

  authored
  43e4ed8

• Update HF conversion script (#110)

  EricHallahan

  authored
  961b410

• Create CITATION.bib (#109)

  StellaAthena

  authored
  0fda83a

Commits on Aug 22, 2021

• fix checkpoint path

  kingoflolz

  committed
  61f8ab4

• add geglu to v2 model

  kingoflolz

  committed
  9d15698

Commits on Aug 21, 2021

• pin tensorflow version in colab demo

  kingoflolz

  committed
  4c61cfa

Commits on Aug 18, 2021

• Adding script to convert fine-tuned sharded weights into pytorch huggingface weights. (#96)

  ablacklama

  authored
  e2f3420

Commits on Aug 13, 2021

• fix Dockerfile

  kingoflolz

  committed
  22de86e

Commits on Aug 12, 2021

• allow restore to smaller pod than trained

  kingoflolz

  committed
  ccce949

Commits on Aug 10, 2021

• Merge remote-tracking branch 'origin/master'

  kingoflolz

  committed
  553a452

• add progress bar to train.py

  kingoflolz

  committed
  485e1bb

• Add W&B lr logging and project naming (#80)

  morganmcg1

  authored
  8aaa4f9

Commits on Aug 9, 2021

• doc update

  kingoflolz

  committed
  a678e30

• Measure grad norm and noise scale in finetuning (#82)

  nostalgebraist

  authored
  1dfdb06

Commits on Aug 8, 2021

• still load ckpt if it lacks an opt (#79)

  nostalgebraist

  authored
  52f6d0b

Commits on Aug 2, 2021

• Fixes for the create_finetune_tfrecords script (#75)

  nostalgebraist

  authored
  386e59d

Commits on Jul 30, 2021

• Add a cleaned-up tfrecord creation script (#74)

  nostalgebraist

  authored
  845c696

Pagination

PreviousNext