Imprint Reader v1.0-0928
This is the step-2,800 Imprint Reader checkpoint used in the Semantic Mount-and-Read Tuning (SMaRT) study. It is based on Qwen3-14B and was trained to describe factual knowledge or behavioral tendencies associated with frozen weight updates mounted in aligned parameter coordinates.
The model is described in the paper Imprint Reader: From Weight-Update Readout to Behavioral Intervention.
Intended use
The model is a research artifact for studying weight-update readout. A readout experiment must supply a candidate update and mount it using the Reader runtime; the unmounted model is not given an update to inspect. This repository contains the merged Reader weights and tokenizer, not the item-specific updates used during training or MetaEdit intervention assets.
The Reader was trained on 17,184 balanced knowledge and behavior items with anchor-free meta-queries. The final training and held-out test inputs, data schema, and Reader runtime are in the companion repository: Readout-Vibe-Alignment.
Limitations
Open-ended readout is not reliable enough for autonomous auditing or safety decisions. A lower target-description loss is a learned compatibility signal, not a guarantee that an update causes the described behavior in the base model. The checkpoint and data should be used for research evaluation.