OpenTome optimizer benchmark — FWE - Transformer++ 1B
-
OpenRaiser/fwe_transformer_1b_adamp_lr1e_3_b1_0_9_b2_0_98_eps_1e_15
1B • Updated -
OpenRaiser/fwe_transformer_1b_adamw_lr1e_3_b1_0_9_b2_0_99_eps_1e_15
1B • Updated -
OpenRaiser/fwe_transformer_1b_adan_lr3e_3_b1_0_9_b2_0_92_b3_0_99_eps_1e_8
1B • Updated -
OpenRaiser/fwe_transformer_1b_apollo_rank512_scale2_channel_std_gap200_lr3e_3_b1_0_9_b2_0_99_eps_1e_12
1B • Updated