© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 42444
    Prev
    Next
    INDEX
    Explanations

    place names and their context

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

    The neuron is triggered by runs of non‐alphanumeric punctuation (especially sequences of tilde characters used as separators or markdown fences).

    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
    ?”
    -2.67
    艫
    -2.59
    甚至是
    -2.47
    
    -2.47
    âncias
    -2.45
    吆
    -2.45
    n
    -2.39
     finds
    -2.38
     éditoriale
    -2.38
    你不是
    -2.34
    POSITIVE LOGITS
     If
    2.91
    </h3>
    2.89
    </em>
    2.48
    но
    2.19
     '',
    2.16
    .“
    2.11
     اطراف
    2.11
    <td>
    2.11
     apparent
    2.09
    のは
    2.00
    Activations Density 0.022%

    No Known Activations