© 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. 7849
    Prev
    Next
    INDEX
    Explanations

    academic references, specify

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

    The neuron activates on numeric tokens or digit sequences (e.g. years, page numbers, decimal values).

    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
     Дэ
    0.96
     społecz
    0.93
    り
    0.93
    ÓN
    0.91
     Андроид
    0.91
    ៖
    0.90
     Гуляць
    0.89
    ﻜ
    0.88
     乀
    0.87
    следования
    0.87
    POSITIVE LOGITS
    '
    0.95
    R
    0.87
    s
    0.84
    p
    0.78
    х
    0.77
    ul
    0.76
    tern
    0.76
    Ther
    0.73
    L
    0.73
    錳
    0.72
    Activations Density 0.002%

    No Known Activations