pullrequests Search Results · bucket language:JavaScript language:JavaScript language:Python language:JavaScript language:Java
Filter by
1M results
... ). 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
ShenJun93
M001NSummary
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
...
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
Uuriko... completing the request. Slow bucket recovery can therefore delay the admin
operation until recovery finishes or times out. Repeated tracker creation failures can keep the request waiting for a ...
Denovo1998
What does this PR do?
This PR adds support for HF Buckets in Trainer. Instead of storing checkpoints locally or on the hub repository, we can
leverage HF Buckets for that (S3 like storage + XET ...
SunMarc
Native capture sends heartbeats synchronously, while the idle collector requests queued bucket creation. A delayed queue
worker could send the first heartbeat before the bucket exists, receive HTTP ...
ahmetcemkaraca
Motivation
TopK v2 can silently select incorrect values when more than 2,048 candidates occupy the threshold bucket. FP32 scores in
a coarse bucket need not be equal, so selecting from an arbitrary ...
jit-kernel
luoroger37