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

    Neuronpedia

    INDEX
    Explanations

    is able, is best, is characterized

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     starts
    0.69
    starts
    0.68
    shells
    0.67
    speople
    0.66
    lara
    0.66
    circles
    0.66
     ensues
    0.65
    cations
    0.64
    ls
    0.63
    gers
    0.63
    POSITIVE LOGITS
     unable
    0.84
     able
    0.82
     difficult
    0.80
    ogonal
    0.77
     characterized
    0.75
     regarded
    0.75
     capaz
    0.75
     classified
    0.74
     emblematic
    0.74
     coated
    0.73
    Act Density 0.000%

    No Known Activations