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

    Neuronpedia

    INDEX
    Explanations

    hyphenated compound words with the first part indicating a continuous or increasing behavior

    oai_token-act-pair · gpt-3.5-turbo

    phrases that describe ongoing changes or increases in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ovie
    -0.72
     Organizations
    -0.72
    OLOG
    -0.71
    itans
    -0.67
    OLOGY
    -0.67
     judgment
    -0.66
    emis
    -0.66
     Skydragon
    -0.66
     judgement
    -0.64
    lasses
    -0.64
    POSITIVE LOGITS
    present
    0.88
    increasing
    0.87
    theless
    0.85
    changing
    0.76
     dreamed
    0.75
     whiff
    0.74
    growing
    0.74
    awaited
    0.73
    con
    0.71
    ce
    0.71
    Act Density 0.027%

    No Known Activations