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

    Neuronpedia

    INDEX
    Explanations

    phrases related to stairs and access to different levels

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Shield
    -0.17
    IPC
    -0.14
    lice
    -0.14
    Å¡ek
    -0.14
     addCriterion
    -0.14
    IFA
    -0.13
    TEE
    -0.13
    tbl
    -0.13
     horizon
    -0.13
     menus
    -0.13
    POSITIVE LOGITS
     stairs
    0.50
     stair
    0.45
     staircase
    0.44
     steps
    0.43
     Steps
    0.37
    steps
    0.35
    stairs
    0.34
    Steps
    0.33
     escal
    0.33
    梯
    0.30
    Act Density 0.054%

    No Known Activations