Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion docs/codex/experiments.md
Original file line number Diff line number Diff line change
Expand Up @@ -176,7 +176,7 @@ region unavailable to tokenizer fitting, training, early stopping, and checkpoin
One-shot best-checkpoint test evaluation confirms ByteBPE512 by `0.0614` BPC on Alice and `0.0258`
on Peter Pan. Those test segments are consumed and must not guide further tuning.

Milestone 027 was preregistered in commit `d4a17dc` before Hamlet corpus access. Under the unchanged
Milestone 027 was preregistered in commit `6f130d5` before Hamlet corpus access. Under the unchanged
seed-1337 sealed protocol, ByteBPE512 beats character by `0.0673` test BPC on the dramatic-play
distribution. Hamlet's terminal segment is easier than validation for both models, so gap direction
and effect magnitude remain corpus-dependent. The Hamlet test is consumed.
2 changes: 1 addition & 1 deletion experiments/027-hamlet-external-distribution.md
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,7 @@ be reported. Test results will not trigger configuration changes or additional H

## Observations

The preregistration above was committed as `d4a17dc89138c9ff6b6271b70a7d04d176a9da5e` at
The preregistration above was committed as `6f130d5b3c9c7b9deaeac09d8e4bb2bd8d325a33` at
`2026-07-13T01:50:48-05:00`, before the first network request. The declared protocol was executed
without substitution or post-access configuration changes.

Expand Down
Loading