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

    Neuronpedia

    INDEX
    Explanations

    references to colleagues and their relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    coworkers or associates

    np_acts-logits-general · gemini-2.0-flash

    colleagues and neighbors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    GenerationType
    -0.37
     Hols
    -0.36
    HSSF
    -0.31
     draw
    -0.30
    Skocz
    -0.29
     itself
    -0.29
     super
    -0.29
     I
    -0.29
     def
    -0.29
     paint
    -0.29
    POSITIVE LOGITS
     colleague
    1.08
     colleagues
    1.06
     collega
    1.05
     Colleagues
    0.98
     teammate
    0.95
     collègues
    0.95
     coworker
    0.93
    mmate
    0.91
     Kollegen
    0.89
     peers
    0.86
    Act Density 0.232%

    No Known Activations