© 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. 15895
    Prev
    Next
    INDEX
    Explanations

    or chosen or chest

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

    The neuron fires most strongly on numeric expressions—especially multi‐digit numbers and 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
    ر
    1.40
    ل
    1.10
    р
    1.05
    ي
    0.97
    悝
    0.93
    न
    0.91
    त
    0.87
    सी
    0.86
    بعة
    0.85
    от
    0.83
    POSITIVE LOGITS
    or
    0.82
    and
    0.81
     اﻷ
    0.81
     łącz
    0.81
     něco
    0.77
     Commune
    0.76
     základě
    0.76
     lá
    0.75
     campeonato
    0.73
     والأ
    0.73
    Activations Density 0.000%

    No Known Activations