--- license: mit tags: - training-data-attribution - influence-functions - interpretability - nanochat ---

STRIDE: Training Data Attribution via Sparse Recovery from Subset Perturbations

Paper Project Page
Trained STRIDE steering operators and the nanochat base checkpoints they attribute, for the four pre-training models in the paper. STRIDE attributes a model's prediction back to the pre-training examples that shaped it, by learning a tiny activation-steering operator instead of retraining. | depth | params | base ckpt step | operator | LDS (Spearman) | |-------|--------|----------------|----------|----------------| | d12 | 286M | 1680 | layer 8, rank 16 | 0.156 | | d16 | 537M | 3584 | layer 10, rank 16 | 0.177 | | d20 | 897M | 3320 | layer 12, rank 16 | 0.158 | | d24 | 1.38B | 5568 | layer 15, rank 16 | 0.165 | ## Layout ``` base_checkpoints//model_.pt # nanochat base checkpoint base_checkpoints//meta_.json # nanochat config operators//operator.pt # trained SteeringOperator state dict operators//subsets.npy # subset membership (K=1000, d=10) operators//meta.json # training config + n_train + operator dims tokenizer/ # shared nanochat tokenizer ``` ## Usage ```python from stride.inference import Stride attr = Stride.from_pretrained("d12") result = attr.attribute(my_queries) ``` The LDS ground-truth losses and the held-out test set live in the dataset repo [`rishitdagli/stride-lds`](https://huggingface.co/datasets/rishitdagli/stride-lds). ## Citation ```bibtex @misc{dagli2026stridetrainingdataattribution, title={STRIDE: Training Data Attribution via Sparse Recovery from Subset Perturbations}, author={Rishit Dagli and Abir Harrasse and Luke Zhang and Florent Draye and Amirali Abdullah and Bernhard Schölkopf and Zhijing Jin}, year={2026}, eprint={2606.05165}, archivePrefix={arXiv}, primaryClass={cs.LG}, url={https://arxiv.org/abs/2606.05165}, } ```