© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15408
    Prev
    Next
    INDEX
    Explanations

    objFile File, k coefficients

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

    This neuron fires on tokens with very large vocabulary‐index IDs—that is, on rare or high‐index subword tokens.

    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
     origem
    0.82
    ERON
    0.80
    eron
    0.77
     reputation
    0.77
    ｄ
    0.77
    ൽ
    0.76
     prolongation
    0.75
     seating
    0.74
     coltiv
    0.74
    ল
    0.74
    POSITIVE LOGITS
    Madam
    0.75
    ngày
    0.71
    documents
    0.71
     Ниж
    0.71
    एनएल
    0.70
    Wür
    0.69
    nimi
    0.69
     Мы
    0.69
    isch
    0.68
    PARA
    0.68
    Activations Density 0.000%

    No Known Activations