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

    Neuronpedia

    INDEX
    Explanations

    phrases starting with What

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     सारा
    0.32
     esistono
    0.32
    Rew
    0.31
     ✔
    0.31
              
    0.31
     néhány
    0.31
               
    0.31
                                
    0.31
     redefine
    0.31
     સા
    0.30
    POSITIVE LOGITS
     happened
    0.85
     exactly
    0.71
     happens
    0.70
     kind
    0.68
    soever
    0.67
     else
    0.65
    exactly
    0.59
     Happened
    0.59
     ELSE
    0.58
     exatamente
    0.57
    Act Density 0.106%

    No Known Activations