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

    Neuronpedia

    INDEX
    Explanations

    watching and listening

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     വരു
    0.40
    ប្រើ
    0.38
    亣
    0.38
    舛
    0.37
     преобразова
    0.37
     ReLU
    0.37
     کشت
    0.37
     процеду
    0.37
     explanation
    0.35
     explanations
    0.35
    POSITIVE LOGITS
     witnessing
    0.96
     watching
    0.95
    watching
    0.90
     Watching
    0.88
     assisting
    0.84
    Watching
    0.84
     cheering
    0.82
     spectator
    0.81
     menyaksikan
    0.81
     Observing
    0.80
    Act Density 0.142%

    No Known Activations