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

    Neuronpedia

    INDEX
    Explanations

    current

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of the current working directory or file path in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ained
    -0.07
     *,
    -0.07
     pairing
    -0.07
    synthesize
    -0.07
    orate
    -0.06
    里的
    -0.06
     defs
    -0.06
    -debug
    -0.06
     fluoride
    -0.06
    583
    -0.06
    POSITIVE LOGITS
     unchanged
    0.07
     moveTo
    0.06
     ;
↵
    0.06
     �
    0.06
     iVar
    0.06
     Queens
    0.06
     compartments
    0.06
     +#+#+#+
    0.06
    FILTER
    0.06
    paths
    0.06
    Act Density 0.020%

    No Known Activations