Language Modeling on LSTM hidden projection
104.66PerplexityDense
Evaluation Results
| Method | Links | |
|---|---|---|
| DenseEvaluation protocol=archived protocol2026.07 | 104.66 | |
| Q diag(s)Evaluation protocol=archived protocol2026.07 | 186.55 | |
| Intrinsic MuonEvaluation protocol=archived protocol2026.07 | 250.67 |