- 17 Feb, 2022 3 commits
-
-
Jared Casper authored
changed all_gather to _all_gather_base in distributed checkpointing See merge request ADLR/megatron-lm!395
-
mshoeybi authored
-
mshoeybi authored
-
- 15 Feb, 2022 1 commit
-
-
Jared Casper authored
vision third phase merge: pretraining methods + mit,swin backbones See merge request ADLR/megatron-lm!384
-
- 11 Feb, 2022 5 commits
-
-
Vijay Korthikanti authored
-
Mohammad Shoeybi authored
Remove specific versions of pytorch, etc. from README so it doesn't go out of date. See merge request ADLR/megatron-lm!392
-
Jared Casper authored
-
Mohammad Shoeybi authored
Adding several things to the text_generation_server that were necessary for the demos See merge request ADLR/megatron-lm!350
-
rprenger authored
-
- 08 Feb, 2022 2 commits
-
-
Jared Casper authored
Standalone embedding stage See merge request ADLR/megatron-lm!385
-
Lawrence McAfee authored
-
- 04 Feb, 2022 3 commits
-
-
Lawrence McAfee authored
-
rprenger authored
-
rprenger authored
-
- 01 Feb, 2022 5 commits
-
-
Lawrence McAfee authored
-
Vijay Korthikanti authored
-
Lawrence McAfee authored
-
Vijay Korthikanti authored
-
Vijay Korthikanti authored
-
- 31 Jan, 2022 2 commits
-
-
Lawrence McAfee authored
-
Lawrence McAfee authored
-
- 29 Jan, 2022 5 commits
-
-
Jared Casper authored
second phase of vision code merge See merge request ADLR/megatron-lm!381
-
Lawrence McAfee authored
-
Vijay Korthikanti authored
-
Jared Casper authored
Combination of several github PRs See merge request ADLR/megatron-lm!383
-
Jared Casper authored
-
- 28 Jan, 2022 9 commits
-
-
-
-
-
-
-
-
-
Vijay Korthikanti authored
-
Vijay Korthikanti authored
-
- 27 Jan, 2022 5 commits
-
-
Vijay Korthikanti authored
-
Lawrence McAfee authored
-
Jared Casper authored
Distributed checkpointing memory fix See merge request ADLR/megatron-lm!379
-
Stas Bekman authored
-
Stas Bekman authored
The paper has this info, so proposing to copy it next to the table. Otherwise it's hard to guess whether you used 40GB A100s or 80GB ones (and secondary, n_gpus per node). Thank you!
-