© 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. 31-GEMMASCOPE-2-RES-16K
    7. 7781
    Prev
    Next
    INDEX
    Explanations

    names starting with M, R, K, or Z

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

    This neuron activates on named entities and brand/model tokens—i.e. proper nouns, personal names, product names, and similar unique identifiers.

    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_31_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
     kalo
    0.71
    jis
    0.68
     mínimo
    0.64
    arom
    0.62
     Anh
    0.61
    ብስብ
    0.61
    ś
    0.59
     न्यूनतम
    0.59
     massima
    0.59
     essayé
    0.59
    POSITIVE LOGITS
    ्ड
    0.88
    влека
    0.83
    ়
    0.80
    apatillas
    0.72
    ォ
    0.71
    ού
    0.68
    า
    0.68
    еты
    0.67
    ים
    0.67
    ARE
    0.66
    Activations Density 0.322%

    No Known Activations