whn09 81aee64ac2 neuron: low-latency completion off by default; rank timeline (DEEPINFER_NEURON_RANK_TIMELINE) and Engram prefault ... The runtime's low-latency completion thread busy-polls the notification queues (~0.7 CPU per process). With 64 ranks on one host the run queue reached ~470 tasks on 192 CPUs at step boundaries; main threads were preempted (Engram gather 3 ms CPU in 15 ms wall), and a different rank paced the collectives every step. NEURON_RT_XU_WORKER_ENABLE_LOW_LATENCY_COMPLETION=0: decode step 580 -> 522 ms, TPOT 204.5 -> 184.2 ms, TTFT 1.60 -> 1.52 s (vllm-sonnet 512, bucket 512), tokens unchanged. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>		2026-10-27 22:24:34 +08:00
..
mtp	neuron: drafter forward as one region, compiled dflash metadata fill; verification + drafts regions on by default	2026-10-27 00:14:32 +08:00
v4_modules	neuron: fp8-activation linears take the NKI kernel (Engram projection); drop unused fp8 hidden copies before region splits; log weights without NKI scales	2026-10-27 11:50:34 +08:00
vision_encoder	initial commit	2026-10-20 12:09:42 +08:00
__init__.py	initial commit	2026-10-20 12:09:42 +08:00
decoder_layer.py	neuron: fp8-activation linears take the NKI kernel (Engram projection); drop unused fp8 hidden copies before region splits; log weights without NKI scales	2026-10-27 11:50:34 +08:00
deepseek_vl.py	Neuron: full DeepSeek-V4.1-Flash load + decode region graphs at EP=64 (issues #21, #25)	2026-10-23 01:39:22 +08:00
deepseek.py	neuron: low-latency completion off by default; rank timeline (DEEPINFER_NEURON_RANK_TIMELINE) and Engram prefault	2026-10-27 22:24:34 +08:00