© 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. 15936
    Prev
    Next
    INDEX
    Explanations

    `in` pairs, `in` ipairs

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

    The neuron fires very strongly on tokens ending in “s” (i.e. plural or ’s‐suffix forms).

    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
     stepwise
    0.73
     Israel
    0.69
     westward
    0.68
    💶
    0.68
     MLS
    0.67
     redistribute
    0.66
     interchange
    0.66
     rewind
    0.66
     Oslo
    0.66
     याचा
    0.66
    POSITIVE LOGITS
     dommages
    0.86
    ures
    0.85
    kert
    0.82
    ak
    0.80
    k
    0.79
    OMBRE
    0.77
    тин
    0.77
    m
    0.77
    ées
    0.75
    ระบบ
    0.74
    Activations Density 0.000%

    No Known Activations