© 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-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73726
    Prev
    Next
    INDEX
    Explanations

    after a period

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

    The neuron strongly activates on punctuation that marks sentence breaks (especially periods and commas).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ә
    -1.30
    ële
    -1.26
    😐
    -1.21
    </>
    
    -1.15
     pedagang
    -1.15
     detik
    -1.14
     what
    -1.13
     Mexiko
    -1.12
     utara
    -1.10
     abstrak
    -1.09
    POSITIVE LOGITS
     there
    1.42
     also
    1.21
    才可以
    1.15
    Its
    1.13
    Sejarah
    1.09
    您可以
    1.09
    もし
    1.06
     włosów
    1.02
     поси
    0.99
    смотра
    0.97
    Activations Density 0.030%

    No Known Activations