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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "droplets"

    oai_token-act-pair · gpt-3.5-turbo

    references to droplets or variations of the word "droplet"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Canary
    -0.80
    ãĥĨãĤ£
    -0.78
     Trident
    -0.75
     specificity
    -0.74
     reconciliation
    -0.73
     Ou
    -0.73
     Clause
    -0.72
    ãĥ¯ãĥ³
    -0.68
    é¾įå¥ij士
    -0.68
     Blessed
    -0.67
    POSITIVE LOGITS
    plet
    1.30
    plets
    1.27
    ppers
    1.15
    pper
    1.05
    ves
    1.03
    ppings
    1.00
    uling
    0.94
    pless
    0.93
    cks
    0.92
    umph
    0.88
    Act Density 0.015%

    No Known Activations