pullrequests Search Results · bucket language:JavaScript language:Java language:Python language:JavaScript language:JavaScript
Filter by
1M results
... relative path and artifact
fingerprint stay the same.
marin-us-east-02a is rejecting writes while its account is capacity suspended. SkyRL previously placed temporary data in
the shared 08a bucket ...
agent-generated
agentic-lint
weaver
loom-oa-dev[bot]
... ). LeakyLenses decays exponentially with e-folding thickness LensDecayLength
(blue). B: Time-mean density (last 50 of a 100 year run) from CFM (Crocus, bucket) with a synthetic forcing at a site
with ...
grinsted
Summary
Add an opt-in bucket size that bounds the non-tensor-parallel gradient all-reduce s buffer.
Symptom Reproduction
- Symptom: On a 64x H200 MoE RL run with TP8 and sequence parallelism ...
guapisolo
ShenJun93
Backport
This will backport the following commits from main to 8.19:
- Skip already indexed buckets on downsample resume (#162670)
!--- Backport version: 9.6.0 --
Questions ?
Please refer to the Backport tool documentation
backport
v8.19.25
gmarouli
Backport
This will backport the following commits from main to 9.4:
- Skip already indexed buckets on downsample resume (#162670)
!--- Backport version: 9.6.0 --
Questions ?
Please refer to the Backport tool documentation
backport
v9.4.10
...
default is zero.
- draft_prefill_use_decode_buckets reuses decode batch buckets for draft-prefill CUDA graphs to reduce retained
attention workspaces. Its default is false.
The changes stay in ...
hy-guo
M001NSmall-M startup discovery can leave larger GEMM buckets untuned. For example, with DeepSeek V4.1 + DSpark, target
prefill does not execute the draft context projections (main_proj and context_wkv ...
chenht2022
Sorts buckets by repository count, largest first, with alphabetical ties. Removes the maintenance section and scripts
pointer from the README, and removes narrow-tag labels from the index and bucket ...
oliver-mee