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

    Neuronpedia

    INDEX
    Explanations

    shift

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on numeric tokens (digits, version numbers, key-combo numbers, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _approx
    -0.07
     хвор
    -0.07
    美国
    -0.07
    268
    -0.06
    ۳۵
    -0.06
    .gold
    -0.06
    .maxcdn
    -0.06
    _TERM
    -0.06
    _distance
    -0.06
    pond
    -0.06
    POSITIVE LOGITS
     sleeve
    0.07
     Nokia
    0.06
     ADM
    0.06
     inher
    0.06
     Cin
    0.06
     uveden
    0.06
    ету
    0.06
    dns
    0.06
     этот
    0.06
     lên
    0.06
    Act Density 0.004%

    No Known Activations