---
library_name: neurovlm
tags:
- neuroimaging
- autoencoder
- brain-maps
---

# NeuroAutoEncoder

Autoencoder weights and brain-mask metadata for NeuroVLM anatomical atlas vectors.

## Files

- `autoencoder.safetensors`: `neurovlm.models.NeuroAutoEncoder` weights.
- `mask.npz`: mask array and affine used to map vectors back to NIfTI space.
- `config.json`: architecture and file metadata for downstream loaders.

Load through `neurovlm.models.load_model("autoencoder")` after the package loader update.

## Revision 2026-10

Weights of the revised NeuroVLM paper. What changed:

- **Deterministic training.** Every stage runs with seeded data loaders and deterministic kernels. Two runs from scratch gave byte-identical weights on the same GPU model, driver and library versions.
- **MLP projector.** The brain-to-text projectors (stages 13 and 14) are residual MLPs (28.3 M parameters) instead of Q-Formers. A package names its projector in `config.json["projector"]`.
- **Stage-14 recipe.** Each training text is projected onto the canonical bank of its own level, every training map is re-binarized per batch, and title tokens weigh 5 in the loss. Five seeds were trained and the one with the lowest validation loss (seed 3) is packaged.
- **Calibration.** Text-to-brain maps are multiplied by one global factor per model, fitted on the validation split and clipped at 1 (`map_calibration.json`, `head` 19.16 and `adapter` 6.13).
- **Title decoding.** After a [NETWORK] or [REGION] tag the decoder writes one of the canonical names in `title_vocabulary.json` (48 network and 83 region titles), so it never invents a title.

`autoencoder.safetensors` is the retrained autoencoder (stage 08). `mask.npz` and `config.json` are unchanged.

| file | size | sha256 |
|---|---|---|
| `autoencoder.safetensors` | 239.7 MB | `151d62eebc805ea79007415355c8b26069af7d9c8ee545b7fddb6a5922731f55` |

Provenance: `scripts/outputs/runs/final` of NeuroVLM (`scripts/pipeline/train_pipeline.py`, finished 2026-10-10T19:56:25), training code at git `51bedf4` with 131 modified files, inputs pinned by `scripts/config/hf_manifest.json` (sha256 `49f3e3a9825b`). `scripts/README.md` records the later runs from scratch that gave the same bytes.

Earlier weights: revision `b7c288b34d68a71dd303a1a9de02c8aadf6dd1ae`.