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

    Neuronpedia

    INDEX
    Explanations

    phrases related to combining or grouping things together

    oai_token-act-pair · gpt-3.5-turbo

    phrases that involve grouping or combining elements into unified structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    gery
    -0.85
    taker
    -0.71
    iron
    -0.70
    ander
    -0.68
    wash
    -0.64
     entertain
    -0.62
    someone
    -0.62
     quoted
    -0.60
     entertained
    -0.59
    gered
    -0.59
    POSITIVE LOGITS
     cohesive
    1.13
     disparate
    1.04
     neatly
    0.98
     overlapping
    0.91
     coherent
    0.89
     orderly
    0.88
     fragmented
    0.87
     separ
    0.86
     cohesion
    0.85
     evenly
    0.85
    Act Density 0.449%

    No Known Activations