© 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. 10254
    Prev
    Next
    INDEX
    Explanations

    section for header, solve it, train effectively, init calculator, center elements, life drama, Calculates heuristic, paced drama

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

    The neuron is strongly activated by numeric tokens (e.g. years, volume/page numbers, question IDs) in citations and metadata.

    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
    EM
    0.79
     Directors
    0.75
    র্
    0.74
    宴
    0.73
     Vishal
    0.72
     Lamps
    0.71
    درا
    0.70
     Chapel
    0.68
     Gary
    0.68
    osz
    0.68
    POSITIVE LOGITS
     Çünkü
    0.73
    laublich
    0.71
    蛄
    0.71
     İç
    0.70
     лишь
    0.70
    ällen
    0.70
    นิด
    0.70
     ascertained
    0.68
     antagonism
    0.68
     إذ
    0.68
    Activations Density 0.000%

    No Known Activations