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

    Neuronpedia

    INDEX
    Explanations

    letter

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Hombre
    0.40
     lungs
    0.38
     formulate
    0.38
    ರಿಂದ
    0.37
     万元
    0.37
     messages
    0.37
     viscous
    0.36
     knuckle
    0.36
    нове
    0.36
    нала
    0.36
    POSITIVE LOGITS
     Letter
    0.73
    Letter
    0.71
    letter
    0.67
     letter
    0.67
     LETTER
    0.65
     Letters
    0.59
    LETTER
    0.55
    Letters
    0.50
    O
    0.50
    kenny
    0.50
    Act Density 0.005%

    No Known Activations