© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 7842
    Prev
    Next
    INDEX
    Explanations

    Win + R, then enter command

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

    The neuron detects sentence‐ending boundaries, i.e. the full stops (periods) marking the ends of sentences.

    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
     chieftain
    0.97
    DEPENDENCIA
    0.89
    व्यक्तित्व
    0.89
     थी
    0.88
     escuelas
    0.87
     Giáo
    0.87
     Autobi
    0.86
     의
    0.85
     chieft
    0.85
     Virgo
    0.84
    POSITIVE LOGITS
    ко
    0.80
    ..\
    0.79
    коло
    0.75
    ِ
    0.75
    licken
    0.70
    ،
    0.70
    begin
    0.69
    注意事项
    0.68
    िनी
    0.68
    ,
    0.68
    Activations Density 0.002%

    No Known Activations