© 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. 7875
    Prev
    Next
    INDEX
    Explanations

    lists of numbers and code

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

    The neuron strongly activates on capitalized tokens—proper names or title‐style words (e.g. product or brand names).

    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
    o
    0.84
     اجمعين
    0.80
    organisation
    0.76
    حيان
    0.75
    ্রয়
    0.75
     числа
    0.74
    ו
    0.74
     Александра
    0.73
    el
    0.73
     organisasi
    0.73
    POSITIVE LOGITS
    ılır
    0.74
    3
    0.71
    网友
    0.70
    sheep
    0.69
     zcela
    0.69
    兰
    0.68
     ৩১
    0.68
    申し上げ
    0.68
    ﻨ
    0.67
    Russ
    0.66
    Activations Density 0.001%

    No Known Activations