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

    Neuronpedia

    INDEX
    Explanations

    pronouns followed by verbs in past tense

    oai_token-act-pair · gpt-3.5-turbo

    references to people expressing their desires or needs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Roose
    -0.68
    aston
    -0.65
    phans
    -0.63
     Craw
    -0.61
    mop
    -0.59
    elia
    -0.58
    haps
    -0.58
    ixels
    -0.58
    Mont
    -0.57
    paragraph
    -0.57
    POSITIVE LOGITS
     wrought
    0.85
    'd
    0.79
     happ
    0.78
    've
    0.77
     happened
    0.76
    self
    0.76
     happen
    0.75
     wanted
    0.73
     preached
    0.73
    're
    0.73
    Act Density 0.168%

    No Known Activations