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

    Neuronpedia

    INDEX
    Explanations

    rewriting text

    np_max-act · gemini-2.0-flash

    The neuron detects user instructions to rephrase or reformulate text (e.g. the words “rephrase,” “reword,” “rewrite,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    По
    -0.07
    čná
    -0.06
     poprvé
    -0.06
    -cr
    -0.06
    hod
    -0.06
     blush
    -0.06
     heal
    -0.06
    ياه
    -0.06
     vary
    -0.06
    記事
    -0.06
    POSITIVE LOGITS
     �
    0.08
    _audit
    0.07
     storytelling
    0.07
    HasMaxLength
    0.07
    ाई
    0.06
    olio
    0.06
    işim
    0.06
     ^=
    0.06
     работы
    0.06
    �
    0.06
    Act Density 0.040%

    No Known Activations