A Byte-Level BPE tokenizer trained on ['arb_Arab', 'ces_Latn', 'cmn_Hani', 'dan_Latn', 'deu_Latn', 'ell_Grek', 'fra_Latn', 'fw_edu', 'hun_Latn', 'ind_Latn', 'ita_Latn', 'jpn_Jpan', 'nld_Latn', 'pol_Latn', 'por_Latn', 'rus_Cyrl', 'spa_Latn', 'swe_Latn', 'tur_Latn', 'vie_Latn'] data from Fineweb-2-HQ.

Training Details

Parameter	Value
Algorithm	Byte-Level BPE
Language	`['arb_Arab', 'ces_Latn', 'cmn_Hani', 'dan_Latn', 'deu_Latn', 'ell_Grek', 'fra_Latn', 'fw_edu', 'hun_Latn', 'ind_Latn', 'ita_Latn', 'jpn_Jpan', 'nld_Latn', 'pol_Latn', 'por_Latn', 'rus_Cyrl', 'spa_Latn', 'swe_Latn', 'tur_Latn', 'vie_Latn']`
Target Vocab Size	42,000
Final Vocab Size	13,977
Pre-tokenizer	custom:addition
Number handling	ltr_3digit
Contraction handling	False
Normalizer	NFC
Special Tokens	`<s>`, `</s>`, `<pad>`, `<unk>`
Training Shards	38

Usage

python

from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("flexitok/maddition_AllL_42000")
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
`yirmi iki+dokuz=otuz bir\ntwenty two+nine=thirty one`	`yirmi, Ġ, iki, +, dokuz, =, otuz, Ġ, bir, \, n, tw, ent, y, Ġ, tw, o, +, n, in`	`1091, 223, 727, 13, 722, 31, 1088, 223, 848, 62, 80, 636, 260, 91, 223, 636, 81, 13, 80, 268`

Command used to create this tokenizer:

['/home/gsa/tokenizers2/flexitok/tokenizer_training/train_tokenizers.py', 'algorithm=bpe', 'vocab_size=42_000', 'langs=[arb_Arab,ces_Latn,cmn_Hani,dan_Latn,deu_Latn,ell_Grek,fra_Latn,fw_edu,hun_Latn,ind_Latn,ita_Latn,jpn_Jpan,nld_Latn,pol_Latn,por_Latn,rus_Cyrl,spa_Latn,swe_Latn,tur_Latn,vie_Latn]', 'data_dir=/scratch/gsa/data/multilingual-addition/', 'output_dir=/scratch/gsa/trained_tokenizers/multilingual_addition', 'pretokenizer=custom:addition', 'number_handling=ltr_3digit', 'add_numbers=false', 'handle_contractions=false', 'unicode_normalization=nfc', 'use_byte_level_regex=false', 'byte_fallback=false', 'strip_zero_width=false', 'cjk_char_split=false', 'add_cjk_chars=false', 'max_lines=-1', 'test_string=yirmi iki+dokuz=otuz bir\\ntwenty two+nine=thirty one', 'hf.publish_to_hf=true', 'hf_repo_prefix=flexitok/', 'hf.hf_repo_id=flexitok/maddition_AllL_42000', 'hf.collections=[flexitok/multilingual_addition_tokenizers]']