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

    Neuronpedia

    INDEX
    Explanations

    references to pointing or attribution of blame

    oai_token-act-pair · gpt-3.5-turbo

    phrases or expressions pointing to specific subjects or claims

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lance
    -0.80
    soever
    -0.74
    hop
    -0.74
    stress
    -0.73
    proxy
    -0.73
    unte
    -0.72
    itialized
    -0.70
    mind
    -0.67
    DH
    -0.67
     operates
    -0.66
    POSITIVE LOGITS
    WARD
    0.68
    TextColor
    0.68
    odan
    0.67
    othy
    0.67
     finger
    0.67
     Genocide
    0.64
    obin
    0.64
     victory
    0.62
     evidence
    0.61
     Victory
    0.61
    Act Density 0.075%

    No Known Activations