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

    Neuronpedia

    INDEX
    Explanations

    phrases concerning ritualized or fantasy narratives with specific characters or scenarios

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

    Follows punctuation marks (.,;!) and implies repetition

    np_acts-logits-general · gemini-2.0-flash

    sentences and clauses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     CWE
    -0.47
    opardy
    -0.45
    warzys
    -0.44
    മ്
    -0.43
     Raub
    -0.42
     recic
    -0.42
    findpost
    -0.41
     sonst
    -0.40
    atier
    -0.40
    Textbox
    -0.40
    POSITIVE LOGITS
    こちらも
    1.70
    Again
    1.45
     Again
    1.41
    again
    1.38
     again
    1.27
    これも
    1.26
     opět
    1.14
    こちらは
    1.07
     AGAIN
    1.02
     опять
    1.00
    Act Density 0.341%

    No Known Activations