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

    Neuronpedia

    INDEX
    Explanations

    longer verbs that are in past tense and describe completed actions

    oai_token-act-pair · gpt-3.5-turbo

    past participles and other forms of verbs that indicate completed actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    cially
    -0.79
    uala
    -0.78
    PAC
    -0.78
    ledge
    -0.76
    Compat
    -0.71
    philis
    -0.69
    姫
    -0.69
     ..........
    -0.68
    ISTORY
    -0.68
    QU
    -0.67
    POSITIVE LOGITS
     portion
    1.19
     portions
    0.90
     contents
    0.86
     entity
    0.86
     party
    0.85
     piece
    0.83
     section
    0.83
     parties
    0.82
     version
    0.80
     object
    0.80
    Act Density 0.266%

    No Known Activations