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

    Neuronpedia

    INDEX
    Explanations

    pull

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on the verb “pull” (especially as part of the phrase “pull off”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    723
    -0.07
    ize
    -0.07
    _desc
    -0.07
    [x
    -0.07
    resents
    -0.07
     encode
    -0.07
    819
    -0.06
    ouis
    -0.06
    　　 　 　 　 　
    -0.06
    izes
    -0.06
    POSITIVE LOGITS
     pull
    0.14
     pulled
    0.13
     pulls
    0.11
     Pull
    0.11
    Pull
    0.11
     pulling
    0.10
    .Pull
    0.10
     �
    0.08
     tug
    0.08
    LL
    0.08
    Act Density 0.012%

    No Known Activations