simarora/cylon-nocumsum-routing-top2-k4-sep-wide-dtype32-p1b-t10b-re0-v02a Updated about 20 hours ago
simarora/m04d26y26-online-hebbian-v10-noSiLU-initSmall-120m-initstd_0.02-featdim_256 Updated 14 days ago
simarora/m04d26y26-online-hebbian-v10-noSiLU-initSmall-120m-initstd_0.02-featdim_22 Updated 14 days ago
simarora/m04d26y26-online-hebbian-v10-noSiLU-initSmall-120m-initstd_0.02-featdim_128 Updated 14 days ago
simarora/m04d22y26-gdn-paper-120m-lr_0.0005-layers_12-dmodel_768-n_2048-head_12-numQK_1 Updated 14 days ago
simarora/m04d18y26-attn-120m-lr_0.0005-layers_12-dmodel_768-n_2048-head_16-baseline Updated 14 days ago
simarora/m04d18y26-attn-120m-lr_0.0005-layers_12-dmodel_768-n_2048-head_12-ffn_4-baseline Updated 14 days ago