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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating importance or comparison

    oai_token-act-pair · gpt-3.5-turbo

    phrases that begin with "one of the" followed by a notable feature, aspect, or opinion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    UME
    -0.81
    cape
    -0.78
    anse
    -0.78
    brace
    -0.77
    iture
    -0.74
    plete
    -0.73
    brance
    -0.72
    ume
    -0.70
    opus
    -0.70
    uild
    -0.69
    POSITIVE LOGITS
     reasons
    1.33
     earliest
    1.23
     ways
    1.11
     strang
    1.11
     easiest
    1.10
     drawbacks
    1.10
     coolest
    1.10
     biggest
    1.08
     hardest
    1.08
     criticisms
    1.06
    Act Density 0.070%

    No Known Activations