Language Modeling on 1 Billion Word Language Modeling Benchmark holdout (test)
28Test Perplexity (10 epochs)
MoE-143M
Updated 4d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2017.01 | 28 | - | |
2017.01 | 31.3 | - | |
2017.01 | 34.1 | - | |
2017.01 | 34.6 | - | |
2017.01 | 34.7 | 30.6 | |
2017.01 | 35.7 | - | |
2017.01 | 36 | - | |
2017.01 | 39.7 | - | |
2017.01 | 44.7 | - | |
2017.01 | 45 | 43.7 | |
2017.01 | 45 | - | |
2017.01 | 45.7 | - | |
2017.01 | 46 | - | |
2017.01 | 46.1 | - | |
2017.01 | - | 67.6 | |
2017.01 | - | 54.1 | |
2017.01 | - | 48.2 |