zils-lite-c5i5
A body for the Zils d3-lite soup pool (Bittensor subnet 127): a fine-tune of Qwen/Qwen3.5-0.8B in the chassis layout, BF16, language weights trained, vision tower from the parent.
Soup fine-tune: the start is the uniform weight average of the two bodies of the 14 November 2026 d3-lite soup (VadymPikulyk/zils-lite-dh2 and qhoenix/zils-lite-decision-avg-abc), both Apache-2.0; credit for the start belongs to their authors and to the authors of the bodies they were built from. Procedure: five LoRA (rank 32, merged) fine-tunes of that average with different seeds, one epoch each at learning rate 5e-5, were averaged uniformly; the published weights are the midpoint (weight 0.5) between that five-seed average and the starting soup average. Training targets were soft: for the written items one quarter the item's own answer and three quarters the answer probabilities of the current d3-mini soup average (teacher distillation from the larger pool's soup: the uniform weight average of Zayaan/zils-mini-hm4, qhoenix/zils-mini-decision-t2avg and joamCriff123/zils-mini-m1113b); for the unlabeled rows the d3-mini soup average's probabilities only.
Training data (about 46,100 rows, one shuffled pass per seed): (1) about 31,100 exam-format decision items of the known task families written with the public Zils authoring recipe (scripts/author_exam.py) by Claude Opus and by a local open model, from train-side splits of the public source datasets, including about 1,100 tool-call items written from rows 90,000-93,000 of the function-call subset of glaiveai/glaive-function-calling-v2 (a dataset without splits that the exam recipe also draws from); (2) 15,000 rows of the train split of tasksource/tasksource-jev-typed-decisions used only as unlabeled distillation inputs (labels not used; sources with commercial or unspecified licence metadata, non-commercial sources excluded). No validator exam items, no public seed exam items and no customer data were used. This repository holds weights only; check the source datasets' terms before reuse.