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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating the source of a piece of text

    oai_token-act-pair · gpt-3.5-turbo

    punctuation, particularly periods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     scoop
    -0.78
     commercially
    -0.76
    ahime
    -0.71
    oscope
    -0.66
     traffickers
    -0.64
     turbines
    -0.63
    zsche
    -0.62
     heel
    -0.62
     suspense
    -0.62
     misfortune
    -0.62
    POSITIVE LOGITS
    ItemTracker
    0.92
     Sheila
    0.86
     Tuls
    0.85
    Assembly
    0.84
    PAC
    0.84
    icio
    0.84
     Patrick
    0.82
    Patrick
    0.80
     Michele
    0.78
     Xavier
    0.78
    Act Density 0.029%

    No Known Activations