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

    Neuronpedia

    INDEX
    Explanations

    Renaissance, Victorian, Medieval eras

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ini
    0.42
    able
    0.40
     clauses
    0.39
     current
    0.39
     thai
    0.38
     Graham
    0.37
     upstream
    0.37
    itemize
    0.37
     han
    0.37
    uin
    0.36
    POSITIVE LOGITS
     era
    1.35
    Era
    1.19
     Era
    1.18
    時代
    1.16
    时代
    1.11
    时期
    1.11
    時代の
    1.11
     시대
    1.06
    时代的
    1.05
     period
    1.03
    Act Density 0.025%

    No Known Activations