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

    Neuronpedia

    INDEX
    Explanations

    expressions of uncertainty or conditional statements

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

    Follows "while I" or "while [pronoun]"

    np_acts-logits-general · gemini-2.0-flash

    while i am

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     even
    -0.76
     inoltre
    -0.73
     either
    -0.72
     invece
    -0.72
    even
    -0.70
    either
    -0.69
    Either
    -0.66
     Either
    -0.66
    elif
    -0.65
     então
    -0.65
    POSITIVE LOGITS
     technically
    1.24
     nominally
    1.03
     itſelf
    0.95
     ostensibly
    0.93
     theoretically
    0.93
     certainly
    0.93
     admittedly
    0.90
    Technically
    0.89
     {},
    0.88
     undoubtedly
    0.87
    Act Density 0.391%

    No Known Activations