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

    Neuronpedia

    INDEX
    Explanations

    direct addresses or references to individuals in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     BrowserModule
    -0.82
    UnsafeEnabled
    -0.76
    ".
    -0.73
    BrowserModule
    -0.73
    .³
    -0.72
    SourceChecksum
    -0.70
    DockStyle
    -0.70
    getDescriptor
    -0.67
    \}\\
    -0.66
     VizieR
    -0.66
    POSITIVE LOGITS
    Attn
    0.63
     addressed
    0.54
    ATTENTION
    0.51
    Attention
    0.49
    ATTN
    0.48
    attention
    0.46
     future
    0.45
    Dear
    0.45
    addressed
    0.44
     whoever
    0.44
    Act Density 0.184%

    No Known Activations