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

    Neuronpedia

    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on occurrences of the word “to” when it appears as a connector in compound or hyphenated expressions (e.g. “balanced-to-unbalanced,” “effector-to-memory,” “excel to python,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    bery
    -0.07
     Framework
    -0.06
    users
    -0.06
    หร
    -0.06
     Buff
    -0.06
     बय
    -0.06
    Profiles
    -0.06
    ιακ
    -0.06
    Bell
    -0.06
    .Sdk
    -0.06
    POSITIVE LOGITS
    -to
    0.08
     targ
    0.07
    to
    0.07
    给
    0.07
    <Props
    0.07
    ETO
    0.06
    	to
    0.06
    alphabet
    0.06
    To
    0.06
     destinationViewController
    0.06
    Act Density 0.029%

    No Known Activations