This dataset contains all model checkpoints acquired while training AE Studio's AESMTE3 submission for the NLB 2021 Challenge. The models are neural-data-transformers and were trained using AE's fork of the neural-data-transformers repo. These model checkpoints are intended to be used by the NLB organizers in order to validate AE's submission.

{"references": ["Pei, Felix, et al., \"Neural Latents Benchmark '21: Evaluating latent variable models of neural population activity,\" 2021, arXiv:2109.04463v4 [cs.LG]"]}