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

    Neuronpedia

    INDEX
    Explanations

    complex

    np_max-act · gemini-2.0-flash

    The neuron never activates on any input and thus does not detect any feature (i.e. it’s a “dead” neuron).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    lists of movies.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     disbelief
    -0.06
     كل
    -0.06
    pwd
    -0.06
    Billy
    -0.06
    一次
    -0.06
    IO
    -0.06
    Pie
    -0.06
    ель
    -0.06
    -app
    -0.06
     compression
    -0.06
    POSITIVE LOGITS
     nuanced
    0.07
    (Packet
    0.07
    .Annotation
    0.06
    ateurs
    0.06
    ματα
    0.06
    .Gray
    0.06
     complexities
    0.06
     reacts
    0.06
     Prot
    0.06
     autob
    0.06
    Act Density 0.025%

    No Known Activations