Dataset Viewer
The dataset could not be loaded because the splits use different data file formats, which is not supported. Read more about the splits configuration. Click for more details.
Couldn't infer the same data file format for all splits. Got {NamedSplit('train'): ('json', {}), NamedSplit('test'): (None, {})}
Error code:   FileFormatMismatchBetweenSplitsError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Fixed Alpha Score-Scale Pilot v1

Public artifacts for a controlled 125M-parameter, 4K-context language-model study of fixed attention score scaling with full sparsemax (alpha=2), full entmax-1.5, and a softmax reference.

Layout

  • metadata/: immutable execution records, plans, logs, source/config snapshots, and SHA-256 manifests.
  • runs/<run-name>/: complete run backups, including configuration, provenance, metrics, final training state, and preregistered analysis checkpoints.
  • diagnostics/: compressed offline attention diagnostics and their manifests.
  • results/: machine-readable aggregate results.

All 15 trajectories use seed 0, 1,280 steps, and 20,971,520 training tokens per run. Eleven belong to the original matrix and four to the preregistered lower-scale boundary extension.

Main result

Score multiplier Entmax-1.5 validation NLL Sparsemax validation NLL
0.0625 5.49297 5.42267
0.125 5.44468 5.41162
0.25 5.40096 5.42571
0.5 5.37384 5.47272
1 5.40024 5.52584
2 5.47426 5.62435
4 5.62522 5.77814

The combined tuned gap is 0.03779 NLL, down 69.92% from 0.12560 at the default scale. Both selected scales are now interior points: c=0.5 for entmax-1.5 and c=0.125 for sparsemax. This remains a single-seed exploratory study whose selection and reporting use the same validation split.

Verification

Each run directory contains backup-manifest.json. Downloaded files should be checked against that manifest before use. The original matrix used repository commit d2fdb0538adec47a95728ce93481aaff56ec19a5; the extension used 69e2a5df64844184a2a144964922b747e039938f. Both have the same frozen training-source hash.

These are exploratory single-seed research artifacts, not production-ready pretrained models. Backend runtime differences must not be attributed to alpha alone because sparsemax and entmax-1.5 use different AdaSplash implementations.

Downloads last month
537