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

    Neuronpedia

    INDEX
    Explanations

    neural network architecture

    np_max-act · gemini-2.0-flash

    This neuron activates on mentions of the “transformer architecture” (i.e., technical references to the model’s underlying neural‐network architecture).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Inbox
    -0.07
     dequeueReusableCell
    -0.06
    725
    -0.06
     searchable
    -0.06
    glas
    -0.06
     setDefaultCloseOperation
    -0.06
     Damon
    -0.06
     Bor
    -0.06
    :''
    -0.06
    “So
    -0.06
    POSITIVE LOGITS
     l�
    0.06
     struggles
    0.06
    ]',
    0.06
    ت
    0.06
     paperwork
    0.06
     stm
    0.06
    0.06
    เวลา
    0.06
     Logging
    0.06
    나
    0.06
    Act Density 0.015%

    No Known Activations