© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9230
    Prev
    Next
    INDEX
    Explanations

    regular expressions and code

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

    This neuron flags the first substantive token at the start of each new segment (i.e. it detects sentence‐ or paragraph‐beginnings).

    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
     photod
    0.81
     Priorities
    0.79
     исход
    0.79
    ু
    0.79
     synonyms
    0.77
    Planned
    0.76
     স্নেহ
    0.76
    лык
    0.75
    сных
    0.75
     Valuation
    0.75
    POSITIVE LOGITS
     館
    0.73
     arrivé
    0.72
    जत
    0.69
     comunque
    0.69
    ürm
    0.69
     correttamente
    0.69
     arrivare
    0.68
     comprend
    0.67
     ώστε
    0.67
     contener
    0.66
    Activations Density 0.000%

    No Known Activations