Inject witness ids into the Megatron forward and train step#1446
Conversation
fzyzcjy
requested review from
Zhichenzzz,
maocheng23,
yueming-yuan and
yushengsu-thu
as code owners
Contributor
|
Warning You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again! |
fzyzcjy
force-pushed
the
tom/pr_chain/trainer_ft/dev_revert_reversed/bracket-megatron-actor-methods-with-the-with-logs-decorator
branch
from
b1f9980 to
01273a5
Compare
fzyzcjy
force-pushed
the
tom/pr_chain/trainer_ft/dev_revert_reversed/inject-witness-ids-into-the-megatron-forward-and-train-step
branch
from
5f53d76 to
3452825
Compare
fzyzcjy
force-pushed
the
tom/pr_chain/trainer_ft/dev_revert_reversed/bracket-megatron-actor-methods-with-the-with-logs-decorator
branch
from
01273a5 to
58e1819
Compare
fzyzcjy
force-pushed
the
tom/pr_chain/trainer_ft/dev_revert_reversed/inject-witness-ids-into-the-megatron-forward-and-train-step
branch
from
3452825 to
fde57ba
Compare
Feeds witness ids through the Megatron forward path so per-sample witness tracing can run: includes witness_ids in the get_batch key lists, splats them into the model forward (forward_only and train_one_step), guards combined-1f1b against witness, and dumps/clears stale witness state after a NORMAL train step via witness_dump_and_clear_stale. The witness_info parameter threading and the witness allocator/module are added with their own features.
fzyzcjy
force-pushed
the
tom/pr_chain/trainer_ft/dev_revert_reversed/bracket-megatron-actor-methods-with-the-with-logs-decorator
branch
from
58e1819 to
46bf61a
Compare
fzyzcjy
force-pushed
the
tom/pr_chain/trainer_ft/dev_revert_reversed/inject-witness-ids-into-the-megatron-forward-and-train-step
branch
from
fde57ba to
a19935b
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Feeds witness ids through the Megatron forward path so per-sample witness tracing can run: includes witness_ids in the get_batch key lists, splats them into the model forward (forward_only and train_one_step), guards combined-1f1b against witness, and dumps/clears stale witness state after a NORMAL train step via witness_dump_and_clear_stale. The witness_info parameter threading and the witness allocator/module are added with their own features.