© 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. 16184
    Prev
    Next
    INDEX
    Explanations

    stated figures, Immediate Environment, main face, IC Design, pre-trained

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

    This neuron activates primarily on numeric tokens (especially decimal numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    words ending in common adverb suffixes like "-ly" or appearing in lists and structured content.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    phrases that function as discourse glue—connectives and degree/emphasis modifiers that link or intensify parts of a sentence.

    oai_token-act-pair · gpt-5Triggered 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
     macam
    0.87
     antiguo
    0.81
     kerja
    0.78
     modelo
    0.77
    めの
    0.77
     vivienda
    0.76
     mañana
    0.75
     virtuales
    0.75
     realidad
    0.75
    cznie
    0.75
    POSITIVE LOGITS
     Предпо
    0.79
    той
    0.78
     형태
    0.75
     정치
    0.75
    摫
    0.75
    ेंद्र
    0.74
    тео
    0.74
    urés
    0.74
     philosophies
    0.72
    बताया
    0.72
    Activations Density 0.000%

    No Known Activations