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

    Neuronpedia

    INDEX
    Explanations

    conjunctions and connectors that indicate relationships or contrasts in ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     there
    -0.15
     alike
    -0.14
    nton
    -0.14
     they
    -0.14
     we
    -0.14
    urn
    -0.13
    assis
    -0.13
    нÑı
    -0.12
     it
    -0.12
    .Null
    -0.12
    POSITIVE LOGITS
     whose
    0.70
    whose
    0.58
     which
    0.56
    which
    0.47
     whom
    0.42
     коÑĤоÑĢÑĭе
    0.40
     who
    0.39
     cui
    0.38
     коÑĤоÑĢÑĭй
    0.38
     który
    0.37
    Act Density 0.185%

    No Known Activations