IBL BWB
Collection
IBL BrainWideBench • 6 items • Updated
NEMO · IBL BrainWideBench
Aligns a unit's waveform and autocorrelogram with a CLIP loss, pretrained on 47,395 IBL Neuropixels units. Five seeds, each with the unit embeddings it produces, for brain-region decoding (TS3). 10.6M parameters per checkpoint.
hf download nerdslab/ibl-bwb-nemo nemo_embeddings_42.pt --local-dir embs
python src/ts3/eval.py probe=mlp emb_path=embs/nemo_embeddings_42.pt
Seeds 42 to 46, as nemo_<seed>.pt and nemo_embeddings_<seed>.pt.
Loading a checkpoint with torch.load needs weights_only=False, since it carries its
run config alongside the weights.
Paper · Benchmark · Docs · Leaderboard · Training runs
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support