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

    Neuronpedia

    INDEX
    Explanations

    phrases related to observation or scrutiny

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing attention or observation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     depended
    -0.79
    anchester
    -0.73
    hers
    -0.71
     ensured
    -0.70
    vernight
    -0.68
    heed
    -0.67
    oux
    -0.66
     insisted
    -0.66
    ochet
    -0.64
    cribed
    -0.62
    POSITIVE LOGITS
     angles
    0.84
     possibilities
    0.80
     positives
    0.77
     datas
    0.76
     magnification
    0.76
     negatives
    0.75
     stimuli
    0.73
     replay
    0.73
     carefully
    0.72
     scoreboard
    0.70
    Act Density 0.266%

    No Known Activations