© 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. 11194
    Prev
    Next
    INDEX
    Explanations

    formal vs casual, improvement

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

    The neuron fires on numeric tokens, especially decimal or floating-point numbers.

    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
     tournament
    0.71
    ↵↵
    0.70
     larg
    0.68
     trident
    0.67
     tiller
    0.67
     lantern
    0.66
     population
    0.65
     leak
    0.64
     pram
    0.63
     Excellency
    0.63
    POSITIVE LOGITS
    ূত
    0.82
    mw
    0.79
    stackoverflow
    0.78
    캬
    0.77
    tudo
    0.76
    m
    0.76
    mse
    0.76
    मध्यप्रदेश
    0.76
    σκευ
    0.75
     beyaz
    0.75
    Activations Density 0.000%

    No Known Activations