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

    Neuronpedia

    INDEX
    Explanations

    reflective character and observation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ス
    0.67
    お
    0.52
    ミ
    0.50
    一
    0.48
    ی
    0.47
    ᾧ
    0.45
    ча
    0.45
    {
    0.42
    ウ
    0.42
     aquela
    0.41
    POSITIVE LOGITS
     konfl
    0.51
    kaar
    0.49
     thinker
    0.47
    obus
    0.47
     prover
    0.47
     Lauf
    0.47
    rmsg
    0.46
     صبر
    0.46
    vern
    0.46
     انصاف
    0.46
    Act Density 0.001%

    No Known Activations