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

    Neuronpedia

    INDEX
    Explanations

    suggestions and attempts

    np_max-act · gemini-2.0-flash

    The neuron responds to words expressing personal intention or effort (e.g. “attempt,” “intent,” “committed,” “believe”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Doch
    -0.07
    Recognition
    -0.07
    .↵↵↵↵↵↵↵↵↵↵↵↵
    -0.06
    Mp
    -0.06
     Mü
    -0.06
     اینتر
    -0.06
    �
    -0.06
    錯
    -0.06
    ambre
    -0.06
    .apps
    -0.06
    POSITIVE LOGITS
     ($.
    0.06
     "#
    0.06
     ČR
    0.06
     üçüncü
    0.06
     Batch
    0.06
    #
    0.06
    ість
    0.06
     lawn
    0.06
     equival
    0.05
     expr
    0.05
    Act Density 0.061%

    No Known Activations