basic

2-Class Softmax Probabilities from Logits Calculator

p_i = exp(z_i) / Σ exp(z_j) for two classes.

Compute the 2-class softmax probabilities (positive vs negative) from raw logits z₁ and z₂: p_i = exp(z_i) / [exp(z₁) + exp(z₂)]. Reports p₁, p₂ (which sum to 1) and the predicted class — used as the final layer of binary neural-network classifiers and multinomial logistic regression.

Daniel Oh By Daniel Oh
Published Last reviewed 1 min read

Inputs

Results

Enter values and click Calculate to see results.
Was this helpful?

How to use this calculator

  1. Fill in the inputs above using the units you already have.
  2. Values update automatically as you type — no submit button needed.
  3. Hover any result row for the underlying formula and intermediate values.

Formula

p_i = exp(z_i) / Σ exp(z_j).

In depth

Compute the 2-class softmax probabilities (positive vs negative) from raw logits z₁ and z₂: p_i = exp(z_i) / [exp(z₁) + exp(z₂)]. Reports p₁, p₂ (which sum to 1) and the predicted class — used as the final layer of binary neural-network classifiers and multinomial logistic regression.