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