© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1628
    Prev
    Next
    INDEX
    Explanations

    sorted list, smaller, small, prolonged

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

    The neuron fires on descriptive adjectives that specify qualities, properties, or types (e.g. statistical, anonymous, commercial, durable).

    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_31_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
     všechny
    0.42
     लिए
    0.40
     Seç
    0.40
     umfang
    0.40
     všetky
    0.39
     G
    0.39
     At
    0.39
     Erişim
    0.38
     ensimmä
    0.38
     különböző
    0.37
    POSITIVE LOGITS
    ,
    0.70
    <unused2206>
    0.57
     راجسټ
    0.52
    ،
    0.51
     कॉनक
    0.50
    ,<
    0.49
     thermoplastic
    0.49
    IZED
    0.48
     yet
    0.48
    ，
    0.48
    Activations Density 0.665%

    No Known Activations