© 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. 5352
    Prev
    Next
    INDEX
    Explanations

    naming or calling something

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

    The neuron selectively fires on content-bearing, high-information tokens (especially nouns and key technical or topical terms) rather than on common function words.

    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
    radas
    0.59
    profiss
    0.56
    emples
    0.55
    姏
    0.55
    рг
    0.55
    ildiği
    0.55
    gte
    0.54
    ncies
    0.54
    >≤</
    0.54
     ২টি
    0.54
    POSITIVE LOGITS
     called
    1.46
     chamado
    1.38
     denominada
    1.34
     denominado
    1.27
     llamado
    1.26
     llamada
    1.26
     chamada
    1.23
     namely
    1.16
     bernama
    1.16
    called
    1.16
    Activations Density 0.876%

    No Known Activations