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

    Neuronpedia

    INDEX
    Explanations

    through text, programming, critical, algorithms, notation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ি
    0.55
    נו
    0.51
    ே
    0.51
    ”،
    0.50
    t
    0.50
    ை
    0.49
    िक
    0.49
    iagn
    0.49
     natthi
    0.48
    ิ
    0.47
    POSITIVE LOGITS
     avenues
    0.59
    途径
    0.53
     clenched
    0.53
    翘
    0.53
     svého
    0.51
     interviews
    0.50
     sheer
    0.50
     meticulous
    0.50
     a
    0.50
     channels
    0.50
    Act Density 0.015%

    No Known Activations