© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129262
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly responds to punctuation markers—especially sentence‐ending periods and commas.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    also providing information

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     our
    -1.77
     we
    -1.22
     Поэтому
    -1.16
     probably
    -1.12
     Our
    -1.10
     hầu
    -1.09
    Our
    -1.09
     often
    -1.02
     багато
    -1.00
    Recent
    -0.99
    POSITIVE LOGITS
     also
    2.08
     также
    1.71
     will
    1.71
     também
    1.46
     también
    1.43
     également
    1.42
    还会
    1.41
     inoltre
    1.38
     También
    1.34
    還會
    1.25
    Activations Density 0.012%

    No Known Activations