external-run:ER-0001
source verifiedBDH-25M-PL
validation loss 1.41random baseline loss 5.6
Checkpoint model card and published run configuration.
Registry note: Published Polish BDH checkpoint and model card.
CodeSOTA · Clickable claim · C-0001Evidence snapshot 2026-08-20
The BDH-25M-PL run reduced validation loss from an approximately 5.6 random baseline to 1.41 after 10,000 steps on 100M Polish byte tokens.
01 · Research context
What this claim answers.
- Hypothesis
- H-0001 · Byte-level loss is an early signal
- Experiment
- E-0001 · BDH-25M-PL training baseline
02 · Evidence ledger
Inspect the rows beneath the sentence.
Current registry values are resolved live when available and compared with the captured claim snapshot. Provenance remains visible when resolution falls back to the snapshot.
03 · Read at the right strength
Verification is not the same as reproduction.
source verified
“CodeSOTA verified” means the evidence row and its source have passed the registry’s verification path. It does not, by itself, mean an independent team reran the protocol. Reproduction is recorded separately in the claim verdict and experiment graph.
04 · Citation record
A stable public object.
- Canonical URL
- https://www.codesota.com/claims/C-0001
- Created by
- Kacper Wikiel
- Created
- 2026-08-20
- Snapshot
- 2026-08-20