CoolFace
Modelpublic

tachiwin/tokenizer_test3

sourceHugging Faceupdated 29d agoView on Hugging Face
0likes
evaluation_summary.json12 linesDownload Raw Back to evaluation
1{2  "catalogue_url": "https://huggingface.co/datasets/tachiwin/language_utils/resolve/main/catalogue.json",3  "languages_in_catalogue": 363,4  "languages_with_samples": 106,5  "samples_evaluated": 327,6  "failed_sample_round_trips": 0,7  "failed_language_round_trips": 0,8  "mean_chars_per_token": 3.313280381440324,9  "median_chars_per_token": 3.048474292306002,10  "min_chars_per_token": 2.114375655823715,11  "max_chars_per_token": 6.93119266055045912}