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

    Neuronpedia

    INDEX
    Explanations

    the names of entities (e.g., songs, people, objects) mentioned in a text

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to names of people, songs, or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    omics
    -0.78
    entimes
    -0.76
    issions
    -0.71
    eties
    -0.70
    ysc
    -0.69
    umes
    -0.69
    ptions
    -0.68
    estyles
    -0.67
    venants
    -0.66
    advertising
    -0.65
    POSITIVE LOGITS
     culprit
    1.15
     perpetrator
    0.89
     protagonist
    0.86
     beneficiary
    0.79
     deceased
    0.79
     antagonist
    0.78
     supplier
    0.77
     person
    0.76
     deity
    0.76
     entity
    0.75
    Act Density 0.143%

    No Known Activations