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

    Neuronpedia

    INDEX
    Explanations

    past tense verbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ot
    0.57
    az
    0.54
    ah
    0.52
    OR
    0.48
    anal
    0.48
    or
    0.48
    IR
    0.47
    añas
    0.46
    ana
    0.46
    UT
    0.46
    POSITIVE LOGITS
     rushed
    0.52
     pondered
    0.51
     cramped
    0.50
     numb
    0.49
     حیر
    0.48
     recently
    0.48
     revisited
    0.48
     ravaged
    0.47
     belonged
    0.47
     gelegt
    0.47
    Act Density 0.003%

    No Known Activations