© 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. 7235
    Prev
    Next
    INDEX
    Explanations

    IC design, cultural memes, internal processes

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

    The neuron is acting as a “number detector,” firing strongly on tokens that are numeric (years, dates, measurements, etc.) rather than on ordinary words.

    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
    рый
    0.99
    рующий
    0.91
     CHIKV
    0.91
    cats
    0.88
    м
    0.88
    reacted
    0.88
    ী
    0.88
    ченные
    0.85
    ченный
    0.85
    ко
    0.85
    POSITIVE LOGITS
    다
    0.69
     collaboration
    0.68
    प्ताह
    0.67
     collaborations
    0.67
     negocios
    0.67
     maravill
    0.66
    ipp
    0.64
     ihnen
    0.64
     profitieren
    0.64
     credible
    0.64
    Activations Density 0.000%

    No Known Activations