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

    Neuronpedia

    INDEX
    Explanations

    times and events that are a bit in the past

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "few" indicating limited quantity or occurrence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    acus
    -0.96
    yrinth
    -0.71
    akings
    -0.70
    Reviewer
    -0.68
    sein
    -0.68
    alam
    -0.66
    roit
    -0.66
    UAL
    -0.65
    webkit
    -0.65
    agn
    -0.65
    POSITIVE LOGITS
     dozen
    1.36
     hundred
    1.22
     thousand
    1.07
    dozen
    1.01
     weeks
    0.97
     months
    0.89
     nights
    0.85
     minutes
    0.85
     days
    0.84
    een
    0.81
    Act Density 0.040%

    No Known Activations