| Benchmark | Model | Rank | Results |
|---|---|---|---|
| language-modelling-on-enwiki8 | Skip Cross-Head Transformer-XL | #23 | Bit per Character (BPC): 1.033Number of params: 41M |
| language-modelling-on-wikitext-103 | Skip Cross-Head Transformer-XL | #48 | Test perplexity: 22.91Validation perplexity: 21.87… |