How Instruction and Reasoning Data shape Post-Training: Data Quality through the Lens of Layer-wise Gradients
Paper
•
2504.10766
•
Published
•
40
Yanhong Li
yanhong-li
AI & ML interests
None yet
Recent Activity
published
a model
8 days ago
yanhong-li/llama3_3b_gdn_v4_hybrid_0_125_S1_LL35_selection
published
a model
8 days ago
yanhong-li/llama3_3b_gdn_v4_hybrid_0_125_S1_LL1_selection
Organizations
None yet
authored
a
paper
6 months ago