flexitok/bpe_ltr_pol_Latn_4000_v2

Byte-Level BPE Tokenizer: pol_Latn (4K)

A Byte-Level BPE tokenizer trained on pol_Latn data from Fineweb-2-HQ.

Training Details

Parameter	Value
Algorithm	Byte-Level BPE
Language	pol_Latn
Target Vocab Size	4,000
Final Vocab Size	5,052
Pre-tokenizer	custom:pol_Latn
Number handling	ltr_3digit
Contraction handling	True
Normalizer	NFC
Special Tokens	<s>, </s>, <pad>, <unk>
Training Shards	2

Usage

from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("flexitok/bpe_ltr_pol_Latn_4000_v2")
tokens = tokenizer.encode("Hello, world!")

Files

• tokenizer.json — Full HuggingFace tokenizer
• vocab.json — Vocabulary mapping
• merges.txt — BPE merge rules

Sample Encoding

Text	Tokens	Token IDs
Hello, world! 12345 This is a test. こんにちは	H, el, lo, ,, Ġw, or, ld, !, Ġ, 123, 45, ĠT, hi, s, Ġis, Ġa, Ġte, st, ., Ġ	42, 743, 384, 14, 262, 338, 2011, 3, 223, 4336, 4398, 397, 1173, 85, 2598, 310, 382, 273, 16, 223