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

    Neuronpedia

    INDEX
    Explanations

    percentages or ratios

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate action or movement towards a specific goal or outcome

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     appropri
    -0.76
    orno
    -0.67
    enforcement
    -0.65
     postage
    -0.64
     mosa
    -0.63
     exch
    -0.62
     enforcement
    -0.62
     coordination
    -0.61
     censored
    -0.60
     censorship
    -0.60
    POSITIVE LOGITS
     earn
    0.96
    venge
    0.91
     finish
    0.89
     save
    0.82
    pload
    0.81
     overcome
    0.81
     win
    0.81
    ggles
    0.81
     celebrate
    0.80
     clinch
    0.77
    Act Density 0.177%

    No Known Activations