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

    Neuronpedia

    INDEX
    Explanations

    dynamic action

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     quietly
    0.42
     آہستہ
    0.42
     unu
    0.40
     quiet
    0.40
     Slightly
    0.40
     постепен
    0.39
    移
    0.39
     постепенно
    0.39
     Quiet
    0.39
     mildly
    0.39
    POSITIVE LOGITS
     erupt
    0.95
     surging
    0.92
     bursting
    0.91
     exploding
    0.91
     burst
    0.90
     erupted
    0.90
     blazing
    0.89
     roared
    0.89
     explodes
    0.86
     explode
    0.86
    Act Density 0.122%

    No Known Activations