flexitok/bpe_ltr_tur_Latn_8000_v2

Byte-Level BPE Tokenizer: tur_Latn (8K)

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

Training Details

Parameter	Value
Algorithm	Byte-Level BPE
Language	tur_Latn
Target Vocab Size	8,000
Final Vocab Size	9,016
Pre-tokenizer	custom:tur_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_tur_Latn_8000_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, orld, !, Ġ, 123, 45, ĠT, his, Ġis, Ġa, Ġtest, ., Ġ, ãģ, ĵ, ã	42, 286, 1272, 14, 1129, 7565, 3, 223, 8949, 3899, 381, 3336, 1109, 302, 1681, 16, 223, 5495, 244, 162