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

    Neuronpedia

    INDEX
    Explanations

    pronouns and auxiliary verbs relating to someone's actions and progress toward a goal

    oai_token-act-pair · gemini-2.0-flash

    invitation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     parte
    -0.87
     part
    -0.85
     PARTE
    -0.77
    Part
    -0.74
    part
    -0.73
     Teil
    -0.69
     PART
    -0.69
     Part
    -0.67
     со
    -0.64
     часть
    -0.60
    POSITIVE LOGITS
     invitation
    1.22
     invites
    1.09
     invite
    1.09
     invited
    1.03
     invitations
    0.96
     invita
    0.94
     asked
    0.91
     convite
    0.91
     convid
    0.90
     invitar
    0.88
    Act Density 0.844%

    No Known Activations