| Parameter | Value |
|---|---|
| Algorithm | Byte-Level BPE |
| Language | arb_Arab |
| Target Vocab Size | 16,000 |
| Final Vocab Size | 16,949 |
| Pre-tokenizer | custom:arb_Arab |
| Number handling | ltr_3digit |
| Contraction handling | True |
| Normalizer | NONE |
| Special Tokens | <s>, </s>, <pad>, <unk> |
| Training Shards | 2 |
1from transformers import AutoTokenizer
2
3tokenizer = AutoTokenizer.from_pretrained("flexitok/bpe_ltr_arb_Arab_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, ,, Ġ, w, orld, !, Ġ, 123, 45, Ġ, This, Ġ, is, Ġ, a, Ġ, t, est | 42, 3848, 81, 14, 223, 89, 10002, 3, 223, 16091, 4208, 223, 12697, 223, 901, 223, 67, 223, 86, 2704 |