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

    Neuronpedia

    INDEX
    Explanations

    animal species

    np_max-act · gemini-2.0-flash

    The neuron activates on common English names of animal groups or taxa (e.g. fish, lizard, reptile, flies).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    "errors
    -0.07
    ubat
    -0.07
     iw
    -0.06
    osex
    -0.06
    texts
    -0.06
    clf
    -0.06
    ertest
    -0.06
    -b
    -0.06
    /release
    -0.06
    ahi
    -0.06
    POSITIVE LOGITS
     holds
    0.08
     slips
    0.07
    CONDITION
    0.07
    operand
    0.07
    620
    0.06
     misses
    0.06
    _INPUT
    0.06
     Hopefully
    0.06
     smooth
    0.06
    Unexpected
    0.06
    Act Density 0.020%

    No Known Activations