summaryrefslogtreecommitdiff
path: root/records/track_10min_16mb
AgeCommit message (Expand)Author
25 hourscommit ttt record (#77)Sam Acquaviva
25 hoursRecord: 10L Mixed Precision: val_bpb=1.2147 (10 layers + int6 middle layers) ...Nan Liu
26 hoursInt6 + MLP 3x + sliding window: val_bpb=1.1574 (#61)Sam Larson
27 hoursRecord: Sliding Window + FP16 Embed + 10L + Muon WD + Overtone Init (val_bpb=...notapplica
27 hoursNew SOTA attempt (#52)spokane-way
27 hoursFix: score final partial window in sliding window eval (#124)Matthew Li
30 hoursAdd record: Sliding Window Eval (stride=64), val_bpb=1.1925 (#50)Matthew Li
30 hoursSOTA attempt (val_bpb=1.2064) (#49)spokane-way
31 hoursfp16 tied embedding + lr/warmdown tuning — val_bpb 1.2197 (#42)Renier Velazco
2 daysLaunch snapshotWill DePue