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

    Neuronpedia

    INDEX
    Explanations

    sentences mentioning contrasting or comparing different aspects or viewpoints

    oai_token-act-pair · gpt-3.5-turbo

    evaluative phrases that discuss the characteristics of something being considered interesting or flawed

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    etsk
    -0.75
    yles
    -0.65
    achus
    -0.65
    llular
    -0.65
     ordered
    -0.62
     vacancies
    -0.59
    gart
    -0.58
     apprentices
    -0.58
    lyn
    -0.57
     suspended
    -0.56
    POSITIVE LOGITS
     standpoint
    1.07
     reasons
    0.92
     sense
    0.87
     Firstly
    0.84
    Firstly
    0.81
     Reasons
    0.81
    cause
    0.78
     perspective
    0.76
    esides
    0.75
    CVE
    0.74
    Act Density 0.943%

    No Known Activations