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

    Neuronpedia

    INDEX
    Explanations

    generative prompts

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    個人
    0.63
    鳶
    0.62
    个人
    0.60
     নিতান্ত
    0.60
    頴
    0.60
     個人
    0.58
     pleaded
    0.57
    䪱
    0.57
     precedence
    0.56
    事前
    0.56
    POSITIVE LOGITS
    Wid
    0.53
    Voids
    0.53
     Volk
    0.53
     Wah
    0.51
     Wid
    0.51
     जेट
    0.51
    Vol
    0.50
    вых
    0.50
    Funding
    0.49
    Pup
    0.49
    Act Density 0.001%

    No Known Activations