© 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. 1005
    Prev
    Next
    INDEX
    Explanations

    the network

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

    The neuron responds to single, concrete nouns denoting physical places or surfaces (e.g. “floor,” “waist,” “world,” “field,” “table”).

    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
    AM
    0.61
    AR
    0.60
    OR
    0.60
    ﺎ
    0.58
    The
    0.57
    の効果
    0.57
    AL
    0.55
    Д
    0.55
    ৰ
    0.55
    entropy
    0.54
    POSITIVE LOGITS
     same
    0.89
     outset
    0.75
     meantime
    0.74
     outskirts
    0.73
    mselves
    0.73
     meisten
    0.73
     vicinity
    0.71
     netherlands
    0.71
     abyss
    0.70
     sidelines
    0.67
    Activations Density 0.649%

    No Known Activations