© 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. 16050
    Prev
    Next
    INDEX
    Explanations

    punctuation and code separators

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

    This neuron flags the start of new sections or topic shifts in the text, spiking on the first salient word of each fresh 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
     voies
    0.85
    nashvillehousing
    0.84
     voir
    0.80
    ibid
    0.78
    llrp
    0.76
    ם
    0.76
    nCurr
    0.76
    Painted
    0.75
    थाम
    0.73
    सूचित
    0.72
    POSITIVE LOGITS
    0.70
     Estadual
    0.69
     doprav
    0.68
     Sieg
    0.67
     daran
    0.66
    ‍♀️
    0.66
     ihe
    0.64
     Erk
    0.64
    </h2>
    0.63
    ^{*}\
    0.63
    Activations Density 0.003%

    No Known Activations