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

    Neuronpedia

    INDEX
    Explanations

    queries or reflective thoughts

    oai_token-act-pair · gpt-3.5-turbo

    reflections and self-directed questions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pour
    -0.78
    nor
    -0.74
    nic
    -0.73
    udi
    -0.68
    von
    -0.67
    edia
    -0.63
     Sale
    -0.63
     Klu
    -0.63
    ga
    -0.63
    resa
    -0.63
    POSITIVE LOGITS
    é¾įå¥ij士
    0.89
    çīĪ
    0.84
    ħĭ
    0.80
    omething
    0.79
     subconscious
    0.74
    terday
    0.73
     puzzled
    0.72
     mate
    0.71
     anew
    0.71
     afterward
    0.70
    Act Density 0.028%

    No Known Activations