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

    Neuronpedia

    INDEX
    Explanations

    complex structures and roles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     moiety
    0.36
     supposedly
    0.36
     tzv
    0.36
     conflic
    0.35
    憎
    0.35
     purportedly
    0.34
     tzw
    0.34
    boten
    0.34
    マンス
    0.33
    据说
    0.32
    POSITIVE LOGITS
     glorified
    1.13
     giant
    0.85
    giant
    0.73
     elaborate
    0.66
    版的
    0.64
     gigantic
    0.64
     mini
    0.60
     miniature
    0.59
     sophisticated
    0.59
     gigantes
    0.59
    Act Density 0.228%

    No Known Activations