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

    Neuronpedia

    INDEX
    Explanations

    phrases that refer to things in the immediate context, emphasizing the word "this"

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

    "this" followed by a function or question

    np_acts-logits-general · gemini-2.0-flash

    this followed by a noun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    httphttps
    -0.60
     riuscito
    -0.58
    grà
    -0.57
    Попис
    -0.56
     Amm
    -0.55
    DrawerToggle
    -0.54
    HasAnnotation
    -0.54
    findpost
    -0.52
    ねて
    -0.51
     dépens
    -0.51
    POSITIVE LOGITS
     <<<<<<<<<<<<<<
    0.69
     stuff
    0.69
    .*")]
    0.59
     TestBed
    0.57
    Tikang
    0.57
     things
    0.56
     thing
    0.56
    DoubleQuotes
    0.56
     as
    0.55
    ,
    0.52
    Act Density 0.202%

    No Known Activations