LoGo: Token-Level Dynamic Local-Global Attention

The ablation conditions require the unavailable implementation, training corpus, checkpoints, and full 1.5B training recipe. Reconstructing only one ablation would not establish the matched controlled comparison.

The report-defining 32k checkpoint, LoGo implementation, and exact benchmark task configurations are unavailable; the fixed dataset registry does not identify these task assets.

The defining 1.5B comparison requires the unavailable LoGo and baseline checkpoints, 100B-token plus two context-extension training stages on the paper's in-house corpus, and an exact lm-evaluation-harness configuration. An independent reconstruction would require material unspecified substitutions and cannot receive strict automatic status.

Strict reproduction requires the author implementation, the in-house pretraining corpus, exact checkpoints or full training, and the complete seed and evaluation configuration; no verified repository or those assets are present in the fixed inputs.

This hardware-sensitive claim needs the missing Triton kernel and dense comparator implementation, exact GPU and software versions, and a complete timing protocol including warmup, timed iterations, synchronization, and input construction. The paper does not determine all of these conditions.

The paper's context-extended checkpoints and exact RULER data/configuration are not provided, and no verified author repository exists. A substitute public model or reduced RULER panel would not cover the reported scope.