Explanation of the word logits in TensorFlow.

Logits are values that are used as a contribution to softmax. To understand this better, snap here, this is true by TensorFlow.

Therefore, Logits are utilized to map probabilities [0,1] to R [-inf, inf]

L=ln(p/1-p) p=1/1+e^-L

From limits, we conclude that the probability approaching 0.5 corresponds to a logit of 0. Hence, positive logits correspond to a probability of greater than 0.5, and negative corresponds to a probability value of less than 0.5.