flexitok/bpe_ltr_hun_Latn_32000_v2

Byte-Level BPE Tokenizer: hun_Latn (32K)

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

Training Details

Parameter	Value
Algorithm	Byte-Level BPE
Language	hun_Latn
Target Vocab Size	32,000
Final Vocab Size	32,745
Pre-tokenizer	custom:hun_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_hun_Latn_32000_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, ello, ,, Ġw, orld, !, Ġ, 123, 45, ĠThis, Ġis, Ġa, Ġtest, ., Ġ, ãģ, ĵ, ãĤ, ĵ, ãģ	42, 11662, 14, 1114, 8071, 3, 223, 16234, 2699, 26600, 363, 261, 860, 16, 223, 9610, 244, 18428, 244, 9610