© 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. 12745
    Prev
    Next
    INDEX
    Explanations

    migration from, lists of sources

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

    The neuron spikes on low-frequency or “unusual” tokens—especially bare numbers, decimals, and rare content words.

    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
     motionProxy
    1.05
     notificações
    0.99
    ಔ
    0.97
     выполнения
    0.93
     commotion
    0.90
     مباراة
    0.89
    ्वती
    0.88
     alteraciones
    0.88
     reembol
    0.88
    <unused68>
    0.87
    POSITIVE LOGITS
    s
    1.02
    0.95
    '
    0.95
    it
    0.92
      
    0.89
    0.88
    ers
    0.88
     It
    0.88
    ↵↵
    0.86
     and
    0.84
    Activations Density 0.000%

    No Known Activations