zils-mini-ft1
A Zils d3-mini soup-pool body (Bittensor subnet 127): the pinned d3-mini chassis layout (Qwen3.5-4B
language and vision tensors, BF16), fine-tuned from
1lock/zils-mini-astrid-m1 at revision b3a18304
(Apache-2.0). It is evaluated on the pool's pinned chassis, which supplies the answer layer, prompt and runtime.
Method
- Start: 1lock/zils-mini-astrid-m1 placed on the pinned chassis with the Zils soup builder
(
zils.soup.build), the way validators grade a single body. - Training: one epoch of LoRA (rank 32, alpha 64, learning rate 5e-5) on the 152 language projections of
the Zils d3 runtime with its fixed answer readout (
zils.teacher train --start), merged into the weights in BF16. No other tensor was trained. - Targets: the verified answer of each item.
Training data
4,992 exam-format decision items written by Claude Opus 5.5 with the Zils exam authoring recipe
(scripts/author_exam.py: one public row as a style example, an independent blind answer, kept only on
agreement), using public training splits only:
- 2,974 items for the 15 current known families: Social IQa, PIQA, SVAMP and GSM8K, GLUE MNLI, RTE, QNLI and STS-B, Crisscrossed Captions (validation), 20 Newsgroups, Glaive function calling v2, dair-ai/emotion, News Headlines Dataset for Sarcasm Detection, and TweetEval sentiment;
- 2,018 items across 99 BIG-bench training tasks (
tasksource/bigbench) and Glaive function calling.
No validator exam items, no public seed items and no customer data were used. This repository holds model weights only, no training text; check each source's terms before reusing it.
Local evaluation (not the validators' exam)
1,191 held-out items, scored with the validators' calibrated Brier skill (soup_grade.evaluate, mean of five
calibration splits): 892 written the same way from rows never used in training, including 15 BIG-bench tasks
kept out of training, plus the 299 current-family items of zils-ai/base-exam-seed-v1 to -v4.
| Start (astrid-m1) | This body | |
|---|---|---|
| All 1,191 items | 71.96 | 74.65 |
| Public seeds as the known block, held-out tasks as the new block | 72.33 | 74.60 |
Validator scores are on a private exam and will differ.
Licence
Apache-2.0, as the start body (1lock/zils-mini-astrid-m1), its bases and Qwen/Qwen3.5-4B. Credit to their authors; please cite BIG-bench (Srivastava et al., 2022) and tasksource (Sileo, 2024) for the training sources.
Model tree for Zayaan/zils-mini-ft1
Base model
Qwen/Qwen3.5-4B-Base