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

    Neuronpedia

    INDEX
    Explanations

    complex sentences, especially those with subordinate clauses and lists

    oai_token-act-pair · gemini-2.0-flash

    conjunctions and punctuation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    хьтан
    -0.56
    :✨
    -0.52
    pecto
    -0.47
     ervan
    -0.46
     typelib
    -0.44
    DockStyle
    -0.44
    زيز
    -0.42
     ouro
    -0.42
     CWE
    -0.42
    انظر
    -0.41
    POSITIVE LOGITS
     all
    1.04
     each
    1.01
     todos
    0.96
     todas
    0.95
    each
    0.91
     allemaal
    0.89
     many
    0.84
     semua
    0.83
     tutte
    0.81
     Each
    0.81
    Act Density 4.441%

    No Known Activations