| moe_coarse_1e19 b16: validation loss at iteration 2 on test set | lm loss value: 3.129897E+00 | lm loss PPL: 2.287163E+01 | | |
| moe_coarse_1e19 b8: validation loss at iteration 2 on test set | lm loss value: 3.129929E+00 | lm loss PPL: 2.287235E+01 | | |
| moe_coarse_1e19 b4: validation loss at iteration 2 on test set | lm loss value: 3.137491E+00 | lm loss PPL: 2.304597E+01 | | |
| moe_coarse_1e19 b3: validation loss at iteration 2 on test set | lm loss value: 3.175067E+00 | lm loss PPL: 2.392843E+01 | | |
| g1_tmoe_coarse_1e19 b16: validation loss at iteration 2 on test set | lm loss value: 3.178994E+00 | lm loss PPL: 2.402257E+01 | | |
| g1_tmoe_coarse_1e19 b8: validation loss at iteration 2 on test set | lm loss value: 3.179076E+00 | lm loss PPL: 2.402455E+01 | | |
| g1_tmoe_coarse_1e19 b4: validation loss at iteration 2 on test set | lm loss value: 3.185608E+00 | lm loss PPL: 2.418198E+01 | | |
| g1_tmoe_coarse_1e19 b3: validation loss at iteration 2 on test set | lm loss value: 3.217265E+00 | lm loss PPL: 2.495977E+01 | | |
| temporal_fine_g3_1e19 b16: validation loss at iteration 2 on test set | lm loss value: 3.171383E+00 | lm loss PPL: 2.384042E+01 | | |
| temporal_fine_g3_1e19 b8: validation loss at iteration 2 on test set | lm loss value: 3.171479E+00 | lm loss PPL: 2.384272E+01 | | |
| temporal_fine_g3_1e19 b4: validation loss at iteration 2 on test set | lm loss value: 3.177317E+00 | lm loss PPL: 2.398233E+01 | | |
| temporal_fine_g3_1e19 b3: validation loss at iteration 2 on test set | lm loss value: 3.205341E+00 | lm loss PPL: 2.466391E+01 | | |
| ALL 1e19 FAKEQUANT DONE | |