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

    Neuronpedia

    INDEX
    Explanations

    breaking down information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Changer
    0.51
    ية
    0.47
     changer
    0.45
    出した
    0.45
    ａ
    0.44
    だい
    0.42
    となり
    0.42
    됐다
    0.42
     dedicate
    0.41
    好評
    0.40
    POSITIVE LOGITS
     হত্যা
    0.45
    ventions
    0.44
    Gonz
    0.44
    jes
    0.42
    MED
    0.42
    MAT
    0.41
    West
    0.41
    lichting
    0.41
     அழுத்த
    0.41
    RA
    0.41
    Act Density 0.014%

    No Known Activations