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

    Neuronpedia

    INDEX
    Explanations

    references to stairs, stairwells, and accessibility features in buildings

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

    stairs, stairwells, or elevators

    np_acts-logits-general · gemini-2.0-flash

    stairs and elevators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Mutagenicity
    -0.66
     Villages
    -0.59
     alman
    -0.54
     Inoc
    -0.53
     villages
    -0.52
     Permalink
    -0.52
     underv
    -0.52
    алу
    -0.51
     Ephe
    -0.50
    tachment
    -0.50
    POSITIVE LOGITS
     stairs
    1.60
     staircase
    1.47
     stair
    1.41
     stairway
    1.39
     Stairs
    1.32
     Stair
    1.23
    Stairs
    1.22
     elevator
    1.21
     escal
    1.19
    Stair
    1.19
    Act Density 0.124%

    No Known Activations