© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 102734
    Prev
    Next
    INDEX
    Explanations

    code comments and separators

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

    comment separator lines (sequences of forward slashes, hashes, or equal signs used as visual dividers in source code).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron responds to long sequences of repeated slash characters (comment-divider lines).

    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
    <i>
    -2.83
    <b>
    -2.69
    '
    -2.58
     &
    -2.56
    ↵↵
    -2.47
    _
    -2.33
    </td>
    -2.27
    </b>
    -2.25
    <sup>
    -2.16
    I
    -2.05
    POSITIVE LOGITS
    ﻒ
    2.73
     This
    2.56
    
    2.45
    苌
    2.28
    StoryboardSegue
    2.27
    不会
    2.25
    ꒥
    2.23
     ,‬
    2.20
    ͎
    2.19
    quelize
    2.17
    Activations Density 0.003%

    No Known Activations