JupiterZhu/T2MLR_135M_lstart13_lend18_10B_FineWebEdu
Text Generation • 0.2B • Updated • 55
T2MLR
T2MLR pretraining checkpoints: 3 scales x 2 token budgets, trained from scratch on FineWeb-Edu.
-
-
JupiterZhu/T2MLR_135M_lstart13_lend18_50B_FineWebEdu
Text Generation • 0.2B • Updated • 59 -
JupiterZhu/T2MLR_362M_lstart9_lend24_10B_FineWebEdu
Text Generation • 0.4B • Updated • 59 -
JupiterZhu/T2MLR_362M_lstart9_lend24_50B_FineWebEdu
Text Generation • 0.4B • Updated • 57 -
JupiterZhu/T2MLR_982M_lstart9_lend24_10B_FineWebEdu
Text Generation • 1B • Updated • 61 -
JupiterZhu/T2MLR_982M_lstart9_lend24_50B_FineWebEdu
Text Generation • 1B • Updated • 58