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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "I" followed by a past-tense verb "d" and associated phrases reflecting the speaker's intentions or experiences

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

    I'd be, I'd do, I'd return

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     niega
    -0.49
    SourceChecksum
    -0.43
    écnicas
    -0.42
    msgTypes
    -0.40
    GEBURTS
    -0.40
    -0.39
    -------------</
    -0.38
     disciplina
    -0.38
    SerializedSize
    -0.38
     estekak
    -0.36
    POSITIVE LOGITS
    RectangleBorder
    0.53
    heng
    0.47
    atrician
    0.45
     jing
    0.44
     grammes
    0.44
    Ecuador
    0.44
     Aene
    0.44
    YEN
    0.43
    Pub
    0.43
    期刊论文
    0.43
    Act Density 0.019%

    No Known Activations