Essential models and datasets used to build the IPDA debate canonical model. Includes ORPO, GRPO iterations, SFT distillation, and golden samples.
AI & ML interests
AI Pluralism; Agentic Communiies; language games
Recent Activity
View all activity
models
69
debaterhub/debate-grpo-iter3-groupD-final
Updated
debaterhub/debate-grpo-iter3-groupD-epoch4
Updated
debaterhub/debate-grpo-iter3-groupD-epoch3
Updated
debaterhub/debate-grpo-iter3-groupD-epoch2
Updated
debaterhub/debate-grpo-iter3-groupD-epoch1
Updated
debaterhub/debate-grpo-iter2-canonical
31B
•
Updated
•
1
debaterhub/debate-grpo-iter2-groupD-lora
Updated
debaterhub/debate-grpo-iter2-groupD-epoch1
Updated
debaterhub/debate-grpo-iter2-groupC-best
Text Generation
•
Updated
•
9
debaterhub/debate-grpo-iter2-groupB-epoch2
Updated
datasets
33
debaterhub/debate-iter2-group-c-grpo
Viewer
•
Updated
•
4.67k
•
4
debaterhub/debate-opus-distilled-group-a
Viewer
•
Updated
•
489
•
20
debaterhub/debate-grpo-group-a
Viewer
•
Updated
•
3.03k
•
14
debaterhub/debate-iter2-rescored
Viewer
•
Updated
•
31.9k
•
31
debaterhub/debate-iter2-synthesis-calls
Viewer
•
Updated
•
174
•
16
debaterhub/debate-iter2-judge-calls
Viewer
•
Updated
•
57
•
15
debaterhub/debate-data-iter2
Viewer
•
Updated
•
7.86k
•
18
debaterhub/ipda-iter2-synthesis-calls
Viewer
•
Updated
•
174
•
18
debaterhub/ipda-iter2-judge-calls
Viewer
•
Updated
•
57
•
18
debaterhub/debate-iter2-judge-synthesis
Updated
•
19