Xingyu Zhu
JupiterZhu
AI & ML interests
None yet
Recent Activity
upvoted a paper about 21 hours ago
Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression submitted a paper about 22 hours ago
Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression authored a paper 1 day ago
Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache CompressionOrganizations
T2MLR pretraining checkpoints: 3 scales x 2 token budgets, trained from scratch on FineWeb-Edu.
-
JupiterZhu/T2MLR_135M_lstart13_lend18_10B_FineWebEdu
Text Generation • 0.2B • Updated • 57 -
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 • 61 -
JupiterZhu/T2MLR_362M_lstart9_lend24_50B_FineWebEdu
Text Generation • 0.4B • Updated • 60
T2MLR pretraining checkpoints: 3 scales x 2 token budgets, trained from scratch on FineWeb-Edu.
-
JupiterZhu/T2MLR_135M_lstart13_lend18_10B_FineWebEdu
Text Generation • 0.2B • Updated • 57 -
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 • 61 -
JupiterZhu/T2MLR_362M_lstart9_lend24_50B_FineWebEdu
Text Generation • 0.4B • Updated • 60