Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    introducing explanations or conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    unfortunately
    0.88
     unfortunately
    0.83
     exclusively
    0.73
     tunnel
    0.71
     tunnels
    0.71
    mainly
    0.70
     sadly
    0.69
    rinos
    0.68
     alas
    0.68
     mostly
    0.67
    POSITIVE LOGITS
     Given
    1.74
    Given
    1.68
     given
    1.49
    given
    1.40
    给定
    1.30
     Consider
    1.27
    Consider
    1.22
     diberikan
    1.18
     किसी
    1.16
    Suppose
    1.14
    Act Density 0.340%

    No Known Activations