Cited claim

Sign in with GitHub
← Publications

Cited claim · P7 · Author-curated

Bielik-Minitron-7B-v3.0-Instruct is the 7B instruction-tuned Bielik v3 model with the Mistral-derived tokenizer, obtained by compressing the 11B variant with structured pruning and knowledge distillation.

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

Structured assertion

Relation: Defined as

term
Bielik-Minitron-7B-v3.0-InstructThe concept this claim defines.
category
instruction-tuned language modelKind of thing the term denotes.
tokenizer
Mistral-derived tokenizerTokenizer of the model.
derivation
compression of the 11B variant by structured pruning and knowledge distillationHow the model was obtained.

Paper citations

arXiv:2604.10799v1 →Revision supplied by author
  1. Section 2, p. 2

    The Bielik Minitron 7B v3 model Kinas et al. [2026] was obtained through compression of the 11B variant rather than being trained independently. This process combined structured pruning with knowledge distillation
  2. Section 5.2, p. 6

    Bielik-Minitron-7B-v3.0-Instruct (original tokenizer) scores 64.09 Kinas et al. [2026].

Concept definitions

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

Bielik-Minitron-7B-v3.0-Instruct

The 7B instruction-tuned Bielik v3 model with the original, Mistral-derived tokenizer, compressed from the 11B variant.

Key bielik_minitron_7b_v3_instruct · version 1047a8c7-2137-4bd8-b506-c9cf1390208f

Defined as

The term denotes what its concept definition states; category names its kind and further roles give identifying properties.

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

Mistral-derived tokenizer

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

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