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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "some" and related phrases indicating certain quantities or unspecified groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    umd
    -0.17
    uria
    -0.15
    ÐIJÑĢÑħÑĸв
    -0.15
    asing
    -0.15
    oga
    -0.15
    848
    -0.15
    iyan
    -0.15
    ndef
    -0.14
    ecture
    -0.14
    uter
    -0.14
    POSITIVE LOGITS
    685
    0.17
    cheon
    0.15
    aris
    0.15
     might
    0.15
    ĭ
    0.15
     may
    0.15
    ettel
    0.14
     however
    0.14
    vern
    0.14
     oblig
    0.14
    Act Density 0.068%

    No Known Activations