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

    Neuronpedia

    INDEX
    Explanations

    kept or retained

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ion
    0.46
    ități
    0.44
     samym
    0.43
     burying
    0.42
    JAVA
    0.42
     প্রথমে
    0.41
     scritt
    0.41
    いの
    0.41
    otras
    0.41
     municipales
    0.40
    POSITIVE LOGITS
    ാള
    0.44
     রঙ্গ
    0.43
     quarters
    0.43
    受到了
    0.41
    OutputPath
    0.40
     outspoken
    0.39
     ঘুরে
    0.39
     이러한
    0.38
    鉄道
    0.38
     sorted
    0.37
    Act Density 0.000%

    No Known Activations