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

    Neuronpedia

    INDEX
    Explanations

    Fix, prevents, adds, corrects

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    $).
    0.41
     pizz
    0.36
    ài
    0.35
     piz
    0.35
    ί
    0.35
    uws
    0.35
     excitement
    0.34
     poisson
    0.34
    clam
    0.34
     SEARCH
    0.33
    POSITIVE LOGITS
    𝙰
    0.38
    এবার
    0.38
    ೇತ್ರ
    0.37
    Astronom
    0.36
     केवाईसी
    0.36
     metallurgy
    0.35
    分析
    0.35
    ए
    0.35
     kriy
    0.35
    repository
    0.35
    Act Density 0.001%

    No Known Activations