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

    Neuronpedia

    INDEX
    Explanations

    sweater, diagnosis, morality

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    addPreferredGap
    0.45
     silvery
    0.44
    ---->
    0.43
     Niv
    0.43
     கடற்க
    0.43
     ----->
    0.42
     analogs
    0.42
     coasts
    0.41
    -->
    0.41
     Alaska
    0.41
    POSITIVE LOGITS
    pt
    0.43
    estan
    0.43
    ogen
    0.40
    指
    0.40
     SP
    0.39
     stitch
    0.39
     bamboo
    0.38
    ...”
    0.38
     ist
    0.38
    BD
    0.38
    Act Density 0.001%

    No Known Activations