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

    Neuronpedia

    INDEX
    Explanations

    sentences about plans, arrangements, and decisions

    oai_token-act-pair · gpt-3.5-turbo

    contractions of "I", "we", and "are"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     citiz
    -0.74
    odied
    -0.67
    ilial
    -0.65
     Kills
    -0.63
     violates
    -0.62
     Virus
    -0.61
     Difference
    -0.61
     distinction
    -0.61
     Fail
    -0.60
     Critics
    -0.60
    POSITIVE LOGITS
     hoping
    1.28
     expecting
    1.24
     anticipating
    1.23
     hopeful
    1.16
     eyeing
    1.16
     confident
    1.14
     aiming
    1.09
     planning
    1.09
     preparing
    1.08
     gearing
    1.08
    Act Density 0.222%

    No Known Activations