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

    Neuronpedia

    INDEX
    Explanations

    instances of personal pronouns and their interactions to indicate storytelling or dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    illance
    -0.16
    eck
    -0.16
    zin
    -0.16
    éry
    -0.15
    nete
    -0.14
    ollapse
    -0.14
    è£ģ
    -0.14
    onga
    -0.14
    _Ptr
    -0.14
    érer
    -0.13
    POSITIVE LOGITS
     talking
    0.59
     talk
    0.59
     talked
    0.56
     speak
    0.56
     spoke
    0.56
     speaking
    0.55
     talks
    0.53
    talk
    0.50
     speaks
    0.48
     spoken
    0.48
    Act Density 0.445%

    No Known Activations