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

    Neuronpedia

    INDEX
    Explanations

    sequences of alphanumeric characters and symbols, particularly those resembling numerical codes or formatting characters

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

    Characters followed by numbers or symbols

    np_acts-logits-general · gemini-2.0-flash

    sequences of uppercase letters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    HtmlAttribute
    -0.49
     conmigo
    -0.38
     habet
    -0.37
    连忙
    -0.36
     comigo
    -0.35
     expressed
    -0.35
    RuleContext
    -0.34
     ingenting
    -0.33
     relying
    -0.32
     fun
    -0.32
    POSITIVE LOGITS
     ब्रेकडाउन
    0.75
     للاسماء
    0.75
     Winaray
    0.74
    Rujuakan
    0.65
    
    0.65
    
    0.64
    хьтан
    0.60
    
    0.59
     الحره
    0.58
    
    0.58
    Act Density 0.396%

    No Known Activations