© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6903
    Prev
    Next
    INDEX
    Explanations

    definition of a term

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

    The neuron strongly activates on relatively long, uncommon content words (e.g. “optimization,” “shipwreck,” “amazing”), i.e. it’s detecting rare/low‐frequency tokens.

    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
    نے
    0.80
    房间
    0.79
    ()`
    0.78
    Doctors
    0.77
    ادیه
    0.77
     号
    0.75
    ب
    0.75
    >`;
    0.73
    Email
    0.71
     とする
    0.71
    POSITIVE LOGITS
     aggregation
    0.84
     aggregator
    0.77
     aggregated
    0.77
     सुषमा
    0.75
     nonnegative
    0.73
     INTE
    0.72
     automation
    0.71
     benzyl
    0.71
     eigenvector
    0.70
     Austro
    0.70
    Activations Density 0.000%

    No Known Activations