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

    Neuronpedia

    INDEX
    Explanations

    statements made by observers or writers reflecting on various issues or topics

    oai_token-act-pair · gpt-3.5-turbo

    verbs that indicate observation or commentary

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rone
    -0.77
    owers
    -0.71
    cele
    -0.70
    Cele
    -0.69
    offic
    -0.68
    ratch
    -0.68
    ugal
    -0.68
    onder
    -0.67
    productive
    -0.66
    inal
    -0.64
    POSITIVE LOGITS
    ometimes
    0.71
    olate
    0.68
    omething
    0.66
     petertodd
    0.65
     herself
    0.64
    :]
    0.64
     Solution
    0.63
    ilver
    0.63
     Parables
    0.63
    olation
    0.62
    Act Density 0.249%

    No Known Activations