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

    Neuronpedia

    INDEX
    Explanations

    references to specific actions or instructions

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

    try the following

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     reunir
    -0.32
     AppCompat
    -0.31
    utilisons
    -0.31
     LoggerFactory
    -0.31
     hjär
    -0.31
    انجليز
    -0.31
     réunions
    -0.30
    pragma
    -0.30
     samfun
    -0.30
    Hauptartikel
    -0.29
    POSITIVE LOGITS
     Try
    1.02
     try
    1.02
    を試
    0.98
    Try
    0.93
     TRY
    0.93
     tries
    0.90
     tried
    0.88
     trial
    0.85
     Tried
    0.84
    試
    0.83
    Act Density 0.040%

    No Known Activations