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

    Neuronpedia

    INDEX
    Explanations

    phrases related to a traumatic and distressing experience, especially involving suffering, deprivation, and confrontation with death

    oai_token-act-pair · gpt-3.5-turbo

    repeated sequences of certain characters or symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     scattering
    -0.76
    odan
    -0.68
     dumping
    -0.67
    lda
    -0.66
     shack
    -0.65
     tremend
    -0.63
     redundancy
    -0.62
     cyan
    -0.61
     deprivation
    -0.61
     comparisons
    -0.60
    POSITIVE LOGITS
    £
    0.98
    ¹
    0.93
    ı
    0.87
    ħ
    0.78
    Whe
    0.78
    į
    0.77
    should
    0.75
    Ĵ
    0.75
    Asia
    0.74
    ¯
    0.74
    Act Density 0.368%

    No Known Activations