© 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. 15983
    Prev
    Next
    INDEX
    Explanations

    give you detailed information

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

    The neuron is strongly activated by the first content word of a sentence or text chunk—i.e. capitalized words that appear at the start of a new segment.

    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
     ludzi
    0.75
     kerana
    0.63
     clás
    0.62
     masti
    0.61
    वैसे
    0.60
    こういう
    0.59
     Vì
    0.59
    ֖
    0.59
    เขา
    0.57
    ायचे
    0.57
    POSITIVE LOGITS
    9
    0.71
    4
    0.68
    lic
    0.65
    2
    0.64
    ologies
    0.64
    aze
    0.62
    ements
    0.59
    0
    0.59
    8
    0.59
    of
    0.57
    Activations Density 0.002%

    No Known Activations