Ettin-encoder RLCD decision models: proper scoring rules, one forward pass, typed answers, stock laya.Agent. Demo Space included.
matt goeckel PRO
mgoeckel
P(doom) <0.001%
AI & ML interests
Building in a cave with a box of scraps
Recent Activity
liked a model 2 days ago
decision-tune/decisiontune-1.0 liked a dataset 6 days ago
choyiny/decidebench liked a dataset 6 days ago
nvidia/Nemotron-PIIOrganizations
None yet
models 5
mgoeckel/oscar-1-400m
Text Classification • 0.4B • Updated • 55 • 4
mgoeckel/oscar-1-150m
Text Classification • 0.2B • Updated • 39 • 1
mgoeckel/oscar-1-68m
Text Classification • 74.8M • Updated • 34 • 1
mgoeckel/oscar-1-32m
Text Classification • 35.7M • Updated • 52 • 1
mgoeckel/oscar-1-17m
Text Classification • 18.5M • Updated • 83 • 1
datasets 0
None public yet