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

    Neuronpedia

    INDEX
    Explanations

    programming constructs such as loops and conditionals

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

    Characters or symbols

    np_acts-logits-general · gemini-2.0-flash

    going backwards

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     further
    -0.54
    further
    -0.51
    Further
    -0.50
     Further
    -0.49
    HasIndex
    -0.48
     FURTHER
    -0.47
    FURTHER
    -0.47
    under
    -0.46
     under
    -0.46
     ISNI
    -0.44
    POSITIVE LOGITS
     backwards
    0.92
    Personensuche
    0.87
     reverse
    0.83
    Попис
    0.82
    [::-
    0.77
    reverse
    0.76
     Reverse
    0.75
    Reverse
    0.75
     backward
    0.74
     reversed
    0.74
    Act Density 0.170%

    No Known Activations