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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the accuracy and existence of information

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

    whether... or

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    almost
    -0.57
     almost
    -0.56
    makeText
    -0.56
     very
    -0.56
     sauf
    -0.55
     muy
    -0.51
    except
    -0.51
    幾乎
    -0.51
    Almost
    -0.50
    MathML
    -0.50
    POSITIVE LOGITS
    それとも
    0.81
     überhaupt
    0.75
     yoksa
    0.72
     anymore
    0.69
     вообще
    0.60
     siquiera
    0.60
     vůbec
    0.59
     quelconque
    0.59
     jemals
    0.54
     depends
    0.54
    Act Density 0.119%

    No Known Activations