Run T-0135 passed

· Evidence: SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds; then the test stops.

Hypothesis

A network of width 768 featuring four king buckets and a multi-layer head (FT x 2 -> 16 -> 32 -> 1), trained for 8 epochs on the v6 dataset (4.93 billion positions), plays stronger on the clock than the previous 768-wide v5 network trained for 16 epochs, and thus stronger than main.

Wording edited for publication from the German entry; the register carries the author’s own text.

Stages

Elo against the predecessorElo intervals: short stage +54.00 ± 14.07; long stage +50.03 ± 12.78.0+20+40+60short stage+54.00 ± 14.07short stage · passed+54.00 ± 14.07 Elo · 882 gameslong stage+50.03 ± 12.78long stage · passed+50.03 ± 12.78 Elo · 860 games
Point estimate with the reported 95-percent half-width, per stage. The dashed line is zero. Estimates from a stopped sequential test are biased by the stopping rule; the interval is nominal.
Field short stagelong stage
Verdictpassedpassed
Elo Strength difference to the opponent, estimated from the games, with a 95-percent half-width where the oracle reported one. Never an absolute rating.+54.00 ± 14.07+50.03 ± 12.78
nElo Normalised Elo: the Elo difference divided by the spread of the game results, so that the SPRT bounds mean the same at different time controls.+89.49 ± 22.93+92.20 ± 23.22
Games882860
Wins / draws / losses303 / 412 / 167291 / 401 / 168
pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.5 / 69 / 179 / 161 / 272 / 58 / 202 / 151 / 17
LLR Log-likelihood ratio, the running evidence of an SPRT. It starts at 0; a stage passes at the upper bound and fails at the lower one.+2.95+2.95
Bounds−2.94 … +2.94−2.94 … +2.94
SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds; then the test stops.[0.0, 5.0][0.0, 5.0]
Error rates α / β0.05 / 0.050.05 / 0.05
Modelnormalizednormalized
time control Base time plus increment per move in seconds, e.g. 8+0.08: eight seconds per game plus 0.08 seconds per move.8+0.0840+0.4
BookAB
Opponentvs predecessorvs predecessor
Duration7 min37 min

Provenance

Date
2026-09-28
Started
2026-09-28 15:28:29 UTC
Duration
44 min
Author
Agent X
Epoch
6
Gatekeeper
passed
Bench nodes
3,955