 Alexander Korolev
|
e0e73cf562
history: cluster-wide outage during Boot D - nodes dark mid-load, investigating recovery
|
1 lună în urmă |
|
Alexander Korolev
|
208a173b0c
history: morning R37 typo investigation (LM-head quant found + disproven as sole cause; linear-backend b12x bisect started)
|
1 lună în urmă |
|
Alexander Korolev
|
4fdabdb105
night shift final: full MXFP4 works on 4x at 131K (158/176, DCP4 blocked by engine bug), R37 = 2.3x prefill champion (3,830 pp) but 150/176 quality; teardown notes
|
1 lună în urmă |
|
Alexander Korolev
|
e2af78e409
night plan interim: full MXFP4 probe booted and passed (51.6 KB/token/rank KV, smoke green); DCP4 interleave fixed to 1
|
1 lună în urmă |
|
Alexander Korolev
|
501e627b2b
night plan: R37 image built+distributed (label 9350429b) and the MXFP4 checkpoint complete on the NAS - both lanes ready, switch needs the go-ahead
|
1 lună în urmă |
|
Alexander Korolev
|
b5e08df460
glm53 full MXFP4: explicit --moe-backend flashinfer_cutlass_afp8 + document the MXFP4 ingestion path
|
1 lună în urmă |
|
Alexander Korolev
|
8433106ff4
night-shift 2026-09-13 plan + R37 (jovian) build profile
|
1 lună în urmă |
