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

    Neuronpedia

    INDEX
    Explanations

    pronouns referring to an unspecified subject

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the pronoun "it" and related phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    venge
    -0.75
    arching
    -0.68
    Reply
    -0.67
    indal
    -0.66
    parts
    -0.65
    its
    -0.63
    ceive
    -0.62
    911
    -0.61
    mini
    -0.61
    arat
    -0.59
    POSITIVE LOGITS
     anecd
    0.71
    eners
    0.62
     irony
    0.59
     strains
    0.59
     lur
    0.58
     beh
    0.58
     pled
    0.58
     apologies
    0.57
     Medline
    0.56
     downside
    0.55
    Act Density 0.537%

    No Known Activations