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

    Neuronpedia

    INDEX
    Explanations

    words related to meetings, interactions, or clashes

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "encounter" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aku
    -0.81
    actor
    -0.79
    direction
    -0.78
    ¶ħ
    -0.78
    roxy
    -0.76
    rogens
    -0.73
    cott
    -0.73
    nesia
    -0.70
    riger
    -0.69
    mington
    -0.69
    POSITIVE LOGITS
    ttes
    0.83
    ãĥķ
    0.83
     encounters
    0.80
     encountered
    0.78
     Garc
    0.73
     Mast
    0.72
     Wander
    0.70
    Dialogue
    0.70
     Metatron
    0.69
     Frazier
    0.69
    Act Density 0.026%

    No Known Activations