JacobiForcing/JacobiForcing_Math_7B_v1
Updated • 5
AI & ML interests
efficient LLM inference
Organization Card
Jacobi Forcing is a new training technique that converts LLMs into native casual parallel decoders. Jacobi Forcing models enjoy up to 4.5x higher tokens-per-forward and 4x wall-clock speedup on coding and math tasks, while retraining near-AR generation quality.
datasets 6
JacobiForcing/OpenThoughts_Math_training_data_n64w32
Viewer • Updated • 230k • 26
JacobiForcing/OpenThoughts_Math_training_data_n16w16
Viewer • Updated • 251k • 27
JacobiForcing/OpenCodeInstruct_training_data_n32w16
Viewer • Updated • 2.16M • 55
JacobiForcing/OpenCodeInstruct_training_data_n16w16
Viewer • Updated • 624k • 38
JacobiForcing/OpenCodeInstruct_length_sorted
Viewer • Updated • 1.64M • 41
JacobiForcing/OpenThought2_length_bucketed
Viewer • Updated • 1.14M • 60