© 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. 112819
    Prev
    Next
    INDEX
    Explanations

    code block comments

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

    code comment directives that disable linting or static analysis rules.

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

    The neuron detects numeric literals (numbers and version‐style digits) in code.

    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
    菹
    -1.40
     貓
    -1.35
     various
    -1.32
    完结
    -1.32
    预约
    -1.31
    Ergebnisse
    -1.30
    他对
    -1.30
    photographe
    -1.29
    アップデート
    -1.26
    了一种
    -1.26
    POSITIVE LOGITS
    4
    1.48
    7
    1.44
    0
    1.41
    3
    1.40
    still
    1.38
    6
    1.35
    8
    1.34
     and
    1.34
    genheiten
    1.24
    1
    1.23
    Activations Density 0.041%

    No Known Activations