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

    Neuronpedia

    INDEX
    Explanations

    sequences of underscores or whitespace characters in code or structured text

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

    return map

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     and
    -0.41
     The
    -0.37
    The
    -0.37
     that
    -0.35
     directly
    -0.33
     the
    -0.33
    ↵↵
    -0.32
    取
    -0.32
    CodedInputStream
    -0.31
     without
    -0.31
    POSITIVE LOGITS
    PerformLayout
    0.69
    нгред
    0.64
     Chwiliwch
    0.60
     €€
    0.57
     tartalomajánló
    0.57
     चीज़ों
    0.57
     stiefe
    0.57
    ſehen
    0.57
    AnchorStyles
    0.57
     mauva
    0.56
    Act Density 0.018%

    No Known Activations