© 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. 12557
    Prev
    Next
    INDEX
    Explanations

    harmful premise or bus

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron flags prominent content words—especially uncommon or “named” nouns (e.g. proper names, distinctive terms, or rare descriptive nouns).

    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
    ഞ്ഞ്
    0.89
     émotion
    0.87
     politica
    0.85
     anormal
    0.84
     política
    0.84
     aumento
    0.84
     cellule
    0.83
     étaient
    0.82
     urination
    0.80
     musculaire
    0.80
    POSITIVE LOGITS
     הת
    0.88
     وعلى
    0.82
     Them
    0.81
    ง
    0.78
     Однако
    0.77
     Hints
    0.77
     Это
    0.75
     الدرا
    0.75
    礎
    0.75
     Ta
    0.75
    Activations Density 0.000%

    No Known Activations