© 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. 4644
    Prev
    Next
    INDEX
    Explanations

    the + noun

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

    This neuron flags the very first word of a new paragraph or section, i.e. paragraph-initial tokens (often a capitalized “The”).

    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
     './
    0.65
     массива
    0.64
    Ко
    0.62
     operable
    0.62
     models
    0.61
     flows
    0.61
     "./
    0.60
     "'
    0.59
     இவற்ற
    0.59
     hydrological
    0.58
    POSITIVE LOGITS
    uomo
    1.01
     ragazzo
    1.00
     Irishman
    0.99
     Englishman
    0.91
    businessman
    0.90
     frontman
    0.90
     capitán
    0.89
     ragazza
    0.88
     Spaniard
    0.88
     giovane
    0.88
    Activations Density 0.148%

    No Known Activations