# Sequence Sorting on Toy Vocabulary Sort 1K sequences (held-out)

Source: [Sequence Sorting on Toy Vocabulary Sort 1K sequences (held-out)](<https://www.sota2.com/research/sota/sequence-sorting-on-toy-vocabulary-sort-1k-sequences-held-out>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-08T07:16:59.224Z

Evaluation of Sequence Sorting performance on the Toy Vocabulary Sort 1K sequences (held-out) dataset, measured using Accuracy, Perplexity, Active Frac α, and TLOps saved.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Sequence Sorting |
| Dataset (including any stated split) | Toy Vocabulary Sort 1K sequences (held-out) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/sequence-sorting-on-toy-vocabulary-sort-1k-sequences-held-out.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Perplexity](<https://www.sota2.com/research/sota/sequence-sorting-on-toy-vocabulary-sort-1k-sequences-held-out.md?metric=Perplexity>): Lower is better; unit: SCALAR.

- [Active Fraction (α)](<https://www.sota2.com/research/sota/sequence-sorting-on-toy-vocabulary-sort-1k-sequences-held-out.md?metric=Active%20Fraction%20(%CE%B1)>): Higher is better; unit: PERCENT.

- [TLOps Saved](<https://www.sota2.com/research/sota/sequence-sorting-on-toy-vocabulary-sort-1k-sequences-held-out.md?metric=TLOps%20Saved>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Baseline | 99.15 | Not recorded | 69fd6e045497b3b670c89370 |
| TSA | 98.78 | Not recorded | 69fd6e045497b3b670c89370 |

### Baseline

Accuracy: 99.15. Result ID: 69fd8d8747f34ce924e02051.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e045497b3b670c89370.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49db1.md>)

[Implementation](<https://github.com/tensorflow/tensor2tensor>)

Recorded evaluation configuration: {"Model Dimension (d)":"128","Number of Layers (L)":"6"}. This record does not assert independent reproduction.

### TSA

Accuracy: 98.78. Result ID: 69fd8d8747f34ce924e02052.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e045497b3b670c89370.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e045497b3b670c89370.md>)

Recorded evaluation configuration: {"Model Dimension (d)":"128","Number of Layers (L)":"6"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.