© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6027
    Prev
    Next
    INDEX
    Explanations

    The neuron detects words that list positive attributes or selling points—terms like “advantage(s),” “benefits,” and “features.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     вопроса
    0.86
    )}$,
    0.84
     affront
    0.84
    ปัญหา
    0.82
     thỏa
    0.80
     frustrate
    0.80
     পুস্তকের
    0.80
     --->
    0.79
    FLICT
    0.79
     médiane
    0.79
    POSITIVE LOGITS
     Advantages
    1.16
     advantages
    1.09
    advantages
    1.09
    Advantages
    1.02
    在于
    0.92
     advantage
    0.89
     Vorteile
    0.89
    ı
    0.87
     ventajas
    0.85
    टर
    0.85
    Activations Density 0.382%

    No Known Activations