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

    Neuronpedia

    INDEX
    Explanations

    elements related to file handling and data processing in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    variable names or identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ch
    -0.38
     is
    -0.36
    ec
    -0.35
    is
    -0.33
     =
    -0.33
    f
    -0.32
    ↵↵
    -0.32
    1
    -0.32
    2
    -0.31
    3
    -0.31
    POSITIVE LOGITS
     ***!
    0.85
    DockStyle
    0.83
     GenerationType
    0.78
    ロウィン
    0.78
    MigrationBuilder
    0.77
    <unused68>
    0.76
    <unused74>
    0.76
    <unused8>
    0.76
    <unused3>
    0.76
    [@BOS@]
    0.76
    Act Density 0.044%

    No Known Activations