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

    Neuronpedia

    INDEX
    Explanations

    expressions of interest or curiosity about future events or developments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     devastated
    -0.69
    Hur
    -0.67
     throats
    -0.65
     Mehran
    -0.64
     throat
    -0.63
     crippling
    -0.63
    except
    -0.62
    essential
    -0.60
     unfit
    -0.60
     FIR
    -0.60
    POSITIVE LOGITS
     speculate
    1.03
     revisit
    0.93
     compare
    0.83
     clarify
    0.82
     rethink
    0.80
     ponder
    0.79
     note
    0.79
     examine
    0.78
     learn
    0.77
     explore
    0.76
    Act Density 0.064%

    No Known Activations