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

    Neuronpedia

    INDEX
    Explanations

    pronouns and verbs indicating possession or action

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "they" and "he", indicating a focus on group dynamics and individual roles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    âĦ¢:
    -0.72
     ®
    -0.68
    ienne
    -0.66
     Occ
    -0.63
     Amid
    -0.63
    DAY
    -0.63
     Sao
    -0.60
    aii
    -0.60
     dstg
    -0.59
    tains
    -0.56
    POSITIVE LOGITS
     mathemat
    0.93
    're
    0.87
    've
    0.83
    'll
    0.83
     glim
    0.81
     ain
    0.79
     blat
    0.77
     gotta
    0.75
     contrace
    0.74
     incent
    0.72
    Act Density 0.321%

    No Known Activations