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

    Neuronpedia

    INDEX
    Explanations

    Tyreal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    0.71
    ون
    0.67
    ми
    0.60
    ed
    0.53
    .
    0.53
    行う
    0.51
    自己的
    0.49
    ある
    0.49
    ForThe
    0.47
    ان
    0.46
    POSITIVE LOGITS
     to
    0.70
    0.65
     is
    0.45
     this
    0.42
     that
    0.41
     С
    0.38
     書い
    0.37
     it
    0.37
     the
    0.36
     their
    0.35
    Act Density 2.583%

    No Known Activations