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

    Neuronpedia

    INDEX
    Explanations

    Helping verbs

    np_max-act · gemini-2.0-flash

    The neuron activates on hedging or speculative language—especially occurrences of modal verbs like “may” or “might.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Dude
    -0.06
     Jim
    -0.06
    Simon
    -0.06
    画
    -0.06
    ука
    -0.06
    458
    -0.06
     林
    -0.06
    翰
    -0.06
     Brewer
    -0.06
     tote
    -0.06
    POSITIVE LOGITS
    ルの
    0.07
    บาล
    0.07
     OCI
    0.06
     pošk
    0.06
     बदल
    0.06
     наблю
    0.06
    verbose
    0.06
    -sum
    0.06
    (Application
    0.06
    ा,
    0.06
    Act Density 0.040%

    No Known Activations