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

    Neuronpedia

    INDEX
    Explanations

    personal and possessive pronouns, and words and phrases associated with them.

    oai_token-act-pair · gemini-2.0-flash

    Actions/Verbs

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     accumulates
    -0.73
     relaxes
    -0.70
     will
    -0.63
    gives
    -0.62
     expands
    -0.61
     می‌
    -0.60
     extends
    -0.60
     would
    -0.59
    would
    -0.59
     collects
    -0.59
    POSITIVE LOGITS
     stepped
    0.68
     managed
    0.68
     ran
    0.68
     dealt
    0.67
    tagHelperRunner
    0.65
     ujednoznacz
    0.64
     opened
    0.64
     turned
    0.63
     stopped
    0.62
     kept
    0.61
    Act Density 3.225%

    No Known Activations