Cited claim

Sign in with GitHub
← Publications

Cited claim · P16 · Author-curated

On the Polish text of the Constitution preamble, APT4 has a fertility ratio of 1.62 tokens per word and the Mistral-derived tokenizer 3.22.

Published by @stw2 · 2026-09-14 · Sources, measurements and interpretation are supplied by the author.

Structured assertion

Relation: Metric comparison

metric
Fertility ratioQuantity compared.
evaluation text
Constitution preambleText the metric is computed on.
language
PolishLanguage of the text.
subject
APT4 tokenizerModel or tokenizer after the tokenizer change.
subject value
1.62 tokens per wordValue of the metric for the subject.
comparator
Mistral-derived tokenizerModel or tokenizer before the tokenizer change.
comparator value
3.22 tokens per wordValue of the metric for the comparator.

Paper citations

arXiv:2604.10799v1 →Revision supplied by author
  1. Table 1, p. 3, Polish TpW column, rows APT4 (1.62) and Bielik 11B v3 (3.22)

  2. Section 7, p. 15

    Despite keeping a comparable vocabulary size (∼32,000 tokens), this change reduces the fertility ratio from 3.22 to 1.62 tokens per word on representative Polish text (Table 1), nearly doubling the effective Polish context capacity.

Concept definitions

Reuse the defining version and key when the meaning fits your assertion.

Metric comparison

The subject and the comparator take subject_value and comparator_value of the metric; benchmark, scope, evaluation_text, language and setting state what the values were measured on, where given. Values may come from separate evaluation runs.

Key metric_comparison · version 2d581473-7212-4ea2-bf80-f0c4b8cb247e

Mistral-derived tokenizer

Tokenizer of the original Bielik v3 models, derived from Mistral's.

Key mistral_tokenizer · version bd3d8329-e626-4805-9fff-f83f146769a7

Constitution preamble

Preamble of the Constitution of the Republic of Poland, in Polish and in its official English translation.

Key constitution_preamble · version 86acabda-0237-47be-8e26-a81500c184aa

Fertility ratio

Average number of tokens required to represent a text.

Key fertility_ratio · version f3b88c39-b184-4cb7-85e6-916f260ad0ad

APT4 tokenizer

Polish-optimised tokenizer of the Bielik v3 PL models.

Key apt4 · version d497f94d-5373-4652-887c-55c001b6472c