| Parameter | Value |
|---|---|
| Algorithm | Byte-Level BPE |
| Language | jpn_Jpan |
| Target Vocab Size | 16,000 |
| Final Vocab Size | 17,431 |
| Pre-tokenizer | custom:jpn_Jpan |
| Number handling | ltr_3digit |
| Contraction handling | True |
| Normalizer | NFC |
| Special Tokens | <s>, </s>, <pad>, <unk> |
| Training Shards | 2 |
1from transformers import AutoTokenizer
2
3tokenizer = AutoTokenizer.from_pretrained("flexitok/bpe_ltr_jpn_Jpan_16000_v2")
4tokens = tokenizer.encode("Hello, world!")tokenizer.json — Full HuggingFace tokenizervocab.json — Vocabulary mappingmerges.txt — BPE merge rules| Text | Tokens | Token IDs |
|---|---|---|
Hello, world! 12345 This is a test. こんにちは | H, ell, o, ,, Ġw, orld, !, Ġ, 123, 45, ĠTh, is, Ġis, Ġa, Ġt, est, ., Ġãģĵ, ãĤĵãģ«, ãģ¡ãģ¯ | 42, 4980, 81, 14, 2117, 9832, 3, 223, 17353, 4690, 10378, 1257, 4909, 2412, 1086, 3761, 16, 4547, 5934, 6901 |