Character-level Language Modeling on Shakespeare modern
55.63AccuracySGHMC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SGHMCStep size=0.000200, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.63 | 4.2601 | |
| pSMILEStep size=0.000200, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.62 | 4.3047 | |
| SMILEStep size=0.000200, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.38 | 4.3045 | |
| pSMILEStep size=0.000020, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.31 | 4.354 | |
| pSMILEStep size=0.002000, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.3 | 4.4454 | |
| SMILEStep size=0.002000, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.27 | 4.8501 | |
| SMILEStep size=0.000020, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.26 | 4.298 | |
| SMILEStep size=0.000002, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.08 | 4.3724 | |
| SGHMCStep size=0.000020, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 55.06 | 4.3427 | |
| pSMILEStep size=0.000002, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.96 | 4.3475 | |
| SGHMCStep size=0.000002, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.95 | 4.371 | |
| SGHMCStep size=0.002000, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.7 | 4.4194 | |
| pSMILEStep size=0.000002, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.64 | 4.4744 | |
| SMILEStep size=0.000200, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.58 | 4.4663 | |
| SMILEStep size=0.002000, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.58 | 4.418 | |
| pSMILEStep size=0.000200, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.57 | 4.4507 | |
| pSMILEStep size=0.000020, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.53 | 4.4928 | |
| SGHMCStep size=0.000020, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.51 | 4.4503 | |
| pSMILEStep size=0.002000, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.48 | 4.4653 | |
| SGHMCStep size=0.000002, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.43 | 4.4855 | |
| SGHMCStep size=0.000200, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 54.41 | 4.717 | |
| SGHMCStep size=0.000200, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.41 | 4.446 | |
| SMILEStep size=0.000020, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.34 | 4.433 | |
| SMILEStep size=0.000002, Evaluation Mode=DE(4), Backbone=NanoGPT (10.8M)2026.02 | 54.28 | 4.4955 | |
| pSMILEStep size=0.000200, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 54.07 | 4.7843 | |
| SMILEStep size=0.000200, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 54 | 4.7828 | |
| SMILEStep size=0.000020, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 53.81 | 4.8393 | |
| pSMILEStep size=0.000020, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 53.76 | 4.8699 | |
| SGHMCStep size=0.000020, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 53.57 | 4.8741 | |
| SMILEStep size=0.000002, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 53.45 | 4.9154 | |
| pSMILEStep size=0.000002, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 53.43 | 4.8855 | |
| SGHMCStep size=0.000002, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 53.37 | 4.9094 | |
| pSMILEStep size=0.002000, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 53.18 | 4.9221 | |
| SMILEStep size=0.002000, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 53.09 | 5.0238 | |
| pSMILEStep size=0.000020, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 53.09 | 5.0828 | |
| SGHMCStep size=0.002000, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 53.02 | 5.0227 | |
| SGHMCStep size=0.000002, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.98 | 5.0693 | |
| SMILEStep size=0.000020, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.96 | 5.0324 | |
| SMILEStep size=0.000200, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.96 | 5.0779 | |
| pSMILEStep size=0.002000, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.94 | 5.0582 | |
| SGHMCStep size=0.000020, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.9 | 5.0547 | |
| pSMILEStep size=0.000002, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.9 | 5.0678 | |
| pSMILEStep size=0.000200, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.9 | 5.0374 | |
| SMILEStep size=0.000002, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.89 | 5.0803 | |
| SGHMCStep size=0.000200, Evaluation Mode=DNN, Backbone=NanoGPT (10.8M)2026.02 | 52.82 | 5.0717 | |
| SMILEStep size=0.002000, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 49 | 5.6973 | |
| SGHMCStep size=0.002000, Evaluation Mode=BDE(1), Backbone=NanoGPT (10.8M)2026.02 | 24.72 | 22.0208 | |
| SGHMCStep size=0.002000, Evaluation Mode=BDE(4), Backbone=NanoGPT (10.8M)2026.02 | 15.29 | 10.9048 |