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

    Neuronpedia

    INDEX
    Explanations

    phrases spoken in dialogues or quotes

    oai_token-act-pair · gpt-3.5-turbo

    dialogues and quotes in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ballot
    -0.70
     Recommend
    -0.69
    ebus
    -0.68
     priority
    -0.68
    jri
    -0.67
     Prediction
    -0.67
    intent
    -0.66
    wcsstore
    -0.64
    idity
    -0.64
    acea
    -0.62
    POSITIVE LOGITS
     recalls
    1.76
     recalled
    1.60
     recounted
    1.58
     recounts
    1.53
     remembers
    1.51
     remembered
    1.18
     historian
    1.17
     recol
    1.17
     recollection
    1.16
     recalling
    1.13
    Act Density 0.469%

    No Known Activations