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

    Neuronpedia

    INDEX
    Explanations

    actions and reducer definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Lateral
    0.49
     dorso
    0.47
    Kwa
    0.47
    Pir
    0.46
    Bottle
    0.46
    idefinite
    0.44
    冢
    0.44
    Lon
    0.43
    Personality
    0.43
    Thermal
    0.43
    POSITIVE LOGITS
     action
    1.11
     actions
    1.04
     reducer
    1.04
     Action
    1.02
    reducer
    0.93
     Actions
    0.93
     ACTION
    0.92
     ACTIONS
    0.92
     acción
    0.89
     acciones
    0.88
    Act Density 0.005%

    No Known Activations