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

    Neuronpedia

    INDEX
    Explanations

    instances of prompts or markers indicating the beginning of significant sections in a document

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

    Text following colons, numbers, or symbols

    np_acts-logits-general · gemini-2.0-flash

    questions and answers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ^(@)
    -0.82
    ſelf
    -0.80
     ſy
    -0.79
     ་་
    -0.79
     itſelf
    -0.78
     raiſ
    -0.76
     iſt
    -0.72
    UIControlState
    -0.71
     doubtnut
    -0.71
     }}$}
    -0.71
    POSITIVE LOGITS
    <eos>
    0.70
    Source
    0.65
    productivity
    0.62
     propOrder
    0.59
    ↵↵
    0.57
     #
    0.56
     @
    0.55
     prioritise
    0.53
     fanbase
    0.52
     hilos
    0.52
    Act Density 0.103%

    No Known Activations