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

    Neuronpedia

    INDEX
    Explanations

    packing, growing, access token

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     theorems
    0.46
     wheels
    0.45
     words
    0.44
    ان
    0.43
     Senate
    0.42
     fillets
    0.41
     humanities
    0.41
    G
    0.41
    W
    0.41
     inflation
    0.41
    POSITIVE LOGITS
    larının
    0.55
    descripcion
    0.55
    rasında
    0.53
     porówn
    0.52
    பட்ச
    0.51
    ności
    0.50
    ensureEqual
    0.50
    żu
    0.49
    ší
    0.49
    stylers
    0.49
    Act Density 0.001%

    No Known Activations