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

    Neuronpedia

    INDEX
    Explanations

    phrases related to things working well together

    oai_token-act-pair · gpt-3.5-turbo

    expressions of compatibility or suitability in relationships and contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ithing
    -0.68
    oldown
    -0.65
    ewitness
    -0.64
    ibi
    -0.62
    ittance
    -0.59
    uploads
    -0.58
    tein
    -0.57
    059
    -0.57
    acco
    -0.56
    ittle
    -0.55
    POSITIVE LOGITS
     well
    1.95
     nicely
    1.80
     perfectly
    1.68
     beautifully
    1.63
     poorly
    1.62
     brilliantly
    1.53
     better
    1.52
     wonderfully
    1.50
    well
    1.49
     adequately
    1.44
    Act Density 0.822%

    No Known Activations