© 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. 11248
    Prev
    Next
    INDEX
    Explanations

    equations, lists, ratings, revenue

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

    The neuron fires strongly on uncommon or out‐of‐vocabulary tokens—i.e. rare or unexpected 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
     इंतजार
    0.73
    嵌
    0.71
    scattering
    0.70
     указа
    0.68
    wot
    0.68
     jednego
    0.67
    ufer
    0.66
    することで
    0.66
    ámenes
    0.66
     एकमात्र
    0.65
    POSITIVE LOGITS
    ת
    1.03
    গত
    0.86
    ה
    0.84
    ت
    0.80
     cliquez
    0.80
     cliquer
    0.78
    yani
    0.77
    NOWLEDGE
    0.76
     médec
    0.76
     dlg
    0.75
    Activations Density 0.001%

    No Known Activations