| Parameter | Value |
|---|---|
| Algorithm | Byte-Level BPE |
| Language | fw_edu |
| Target Vocab Size | 4,000 |
| Final Vocab Size | 5,048 |
| Pre-tokenizer | custom:fw_edu |
| 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_fw_edu_4000_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, ,, Ġworld, !, Ġ, 123, 45, ĠThis, Ġis, Ġa, Ġtest, ., Ġ, ã, ģ, ĵ, ã, Ĥ | 42, 477, 81, 14, 904, 3, 223, 4983, 3405, 655, 313, 260, 1074, 16, 223, 162, 226, 244, 162, 227 |