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

    Neuronpedia

    INDEX
    Explanations

    Similarly

    np_max-act · gemini-2.0-flash

    The neuron activates on sentence-initial comparative/discourse adverbs (e.g. “Similarly”) that introduce a parallel or analogous point.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _area
    -0.07
    \Template
    -0.06
     πρά
    -0.06
     Oktober
    -0.06
    FB
    -0.06
     oz
    -0.06
     пр
    -0.06
    �
    -0.06
    _msg
    -0.06
     ortadan
    -0.06
    POSITIVE LOGITS
     Similarly
    0.09
    Similarly
    0.07
     similarly
    0.07
     thrift
    0.07
     Willie
    0.06
     Epoch
    0.06
    ené
    0.06
    ROP
    0.06
    EEP
    0.06
    awe
    0.06
    Act Density 0.009%

    No Known Activations