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

    Neuronpedia

    INDEX
    Explanations

    verbs and forms of "to be" indicating states or conditions

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

    preceding intensifiers or degree modifiers

    np_acts-logits-general · gemini-2.0-flash

    really good/bad

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     poichè
    -0.81
     ainfi
    -0.76
     feroit
    -0.66
     inoltre
    -0.65
    ]}"
    -0.65
     poiché
    -0.63
    几人
    -0.63
    깥
    -0.63
     الرياضيه
    -0.62
     آنان
    -0.62
    POSITIVE LOGITS
     sort
    1.15
     gonna
    1.12
     kind
    1.11
     really
    1.03
     somebody
    0.93
     Yeah
    0.88
     maybe
    0.87
    sort
    0.87
     yeah
    0.86
     very
    0.85
    Act Density 0.135%

    No Known Activations