Article-grounded definition
Symbol C_RL
the compute spent training a policy against a reward model or a verifier, priced in GPU-hours and bounded by the same hardware capability development competes for.
Symbol in context
This entry links every published equation using this exact notation. Its meaning may change between equations.
Meanings in context
Article-grounded definition
Symbol C_RL
the alignment share of.
Used in 4 equations
What Alignment Actually Costs · Equation 1
This equation gives an approximation: it relates the quantities while allowing an approximation.
Equation guide → · This term → · Article →What Alignment Actually Costs · Equation 3
the compute spent training a policy against a reward model or a verifier, priced in GPU-hours and bounded by the same hardware capability development competes for.
Equation guide → · This term → · Article →What Alignment Actually Costs · Equation 11
This equation states an equality: the expressions on both sides have the same value under the article’s assumptions.
Equation guide → · This term → · Article →What Alignment Actually Costs · Equation 15
the alignment share of.
Equation guide → · This term → · Article →