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

    Neuronpedia

    INDEX
    Explanations

    words related to contrasting or opposing statements

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate agreement or concession in an argument

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    erest
    -0.78
    à¼
    -0.75
    aciously
    -0.74
    milo
    -0.73
    Contents
    -0.71
    rap
    -0.70
    zona
    -0.70
    eps
    -0.68
    illus
    -0.68
    playing
    -0.67
    POSITIVE LOGITS
     experts
    1.14
     researchers
    1.13
     analysts
    1.07
     organizers
    1.02
     officials
    1.02
     advocates
    0.97
     investigators
    0.96
     lawmakers
    0.95
     critics
    0.95
     scientists
    0.93
    Act Density 0.471%

    No Known Activations