Advanced
Artificial Intelligence & Machine Learning
Q94 / 100
What is the temperature parameter in LLM sampling?
Correct! Well done.
Incorrect.
The correct answer is B) A parameter controlling randomness in token sampling by scaling logits before softmax — lower temperature → more deterministic, higher → more diverse/random outputs
B
Correct Answer
A parameter controlling randomness in token sampling by scaling logits before softmax — lower temperature → more deterministic, higher → more diverse/random outputs
Explanation
Temperature T: logits/T before softmax. T=0 → greedy (always highest probability token). T=1 → sample from model distribution. T>1 → flatter distribution, more diverse but less coherent. Nucleus (top-p) and top-k sampling offer alternatives.
Progress
94/100