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

    Neuronpedia

    INDEX
    Explanations

    asking for information or code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ꮁ
    0.54
     yalnızca
    0.52
    ᒐ
    0.50
     substant
    0.50
     foundational
    0.49
    ᚄ
    0.47
     whereupon
    0.47
    ადა
    0.47
    Ꮔ
    0.47
     Gardens
    0.47
    POSITIVE LOGITS
     Nowadays
    0.73
     advices
    0.71
     bellow
    0.68
     nowadays
    0.66
     Beside
    0.66
    Nowadays
    0.64
     drawed
    0.64
     loosing
    0.63
    splitted
    0.61
    varage
    0.60
    Act Density 0.180%

    No Known Activations