© 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. 5523
    Prev
    Next
    INDEX
    Explanations

    LeakyReLU, local, threshold, sum

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

    The neuron fires on tokens that describe static states or locations and associated features—e.g. “resides in,” “located in,” “features,” and amenities like “complimentary” or “desks.”

    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
    േരി
    0.67
    镙
    0.62
     đòi
    0.59
    炤
    0.58
     가능
    0.58
    가능
    0.57
    ವೇಂದ್ರ
    0.57
     daughters
    0.56
     संभव
    0.56
     관한
    0.56
    POSITIVE LOGITS
     sécurité
    0.68
    ⌐
    0.67
     fridge
    0.66
     swat
    0.65
     maaf
    0.64
    laughs
    0.63
     anonymity
    0.62
     वॉलेट
    0.61
     quirks
    0.61
     rápidamente
    0.60
    Activations Density 0.075%

    No Known Activations