© 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. 10-GEMMASCOPE-RES-131K
    7. 44510
    Prev
    Next
    INDEX
    Explanations

    conjunctions and punctuation

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

    The neuron fires on low‐content, structural tokens – things like punctuation marks (commas, periods, parentheses), digits, and other common “stop” or function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     این
    -1.88
    Furthermore
    -1.69
    』(
    -1.59
    एक
    -1.59
    éroport
    -1.57
    SenderId
    -1.55
    </h2>
    -1.53
    Moreover
    -1.48
    ❋
    -1.47
    💠
    -1.46
    POSITIVE LOGITS
     they
    1.83
    3
    1.78
     it
    1.60
    0
    1.60
    p
    1.59
    4
    1.55
    1
    1.53
    したので
    1.46
    5
    1.44
    >
    1.39
    Activations Density 0.115%

    No Known Activations