whn09 deb8c5a958 neuron: single-request prompt chunks attend densely over their own SWA ring and compressed blocks, masks from positions (DEEPINFER_NEURON_PREFILL_DENSE, check mode DEEPINFER_NEURON_PREFILL_DENSE_CHECK) ... Within 1 bf16 ulp of the gathered kernel on the full model (mask equality checked on every layer). With a 1024 context bucket: TTFT 1190 -> 1113 ms; tokens differ from the kernel's by rounding (2.89 -> 3.02 tokens a step). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>		2026-10-28 09:09:34 +08:00
..
deep_ep	neuron: Engram gather via numpy (torch index_select waited 12-26 ms on its contended intra-op pool); DEEPINFER_NEURON_TORCH_THREADS; CPU bind mode 'all'	2026-10-28 02:10:43 +08:00
deep_gemm	Neuron: prompt chunks through the decode region graphs (issue #26)	2026-10-23 03:19:56 +08:00
deep_select	Neuron: nkilib rotational_topk for the sampler; single-host shared engram tables (#27)	2026-10-23 07:23:32 +08:00
flash_attn	Neuron basic mode — part 2: deep_ep, sampling kernels, fixes	2026-10-20 18:57:27 +08:00
flash_mla	neuron: single-request prompt chunks attend densely over their own SWA ring and compressed blocks, masks from positions (DEEPINFER_NEURON_PREFILL_DENSE, check mode DEEPINFER_NEURON_PREFILL_DENSE_CHECK)	2026-10-28 09:09:34 +08:00
tile_kernels	Neuron: full DeepSeek-V4.1-Flash load + decode region graphs at EP=64 (issues #21, #25)	2026-10-23 01:39:22 +08:00
tilelang	Neuron: static decode graphs, static-shape MoE, trn2 e4m3 fp8 (issues #9, #12, #24)	2026-10-22 11:59:37 +08:00