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

    Neuronpedia

    INDEX
    Explanations

    professional and legal contexts

    np_max-act · gemini-2.0-flash

    information related to legal expertise and professionalism in various fields.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron detects the boilerplate biographical profile language in professional résumés—e.g. “has,” “is,” “received,” “experience,” “clients,” and similar terms that introduce a person’s background or credentials.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    赫
    -0.07
     Franken
    -0.07
    Undo
    -0.07
     saturated
    -0.06
     survive
    -0.06
    pending
    -0.06
    serial
    -0.06
    optic
    -0.06
     leave
    -0.06
    Т
    -0.06
    POSITIVE LOGITS
    <()>
    0.07
     меся
    0.07
     ();↵↵
    0.07
     t�
    0.07
     AudioSource
    0.07
     spolu
    0.06
    ><!--
    0.06
    ")));↵
    0.06
     pneum
    0.06
     éxito
    0.06
    Act Density 0.020%

    No Known Activations