© 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. 15225
    Prev
    Next
    INDEX
    Explanations

    code delimeters and imports

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

    This neuron fires on the main package or module names in import‐ and using‐style statements (e.g. the “org”, “com”, “io” tokens in imports).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     figuras
    -1.11
    schwar
    -1.07
     tenisky
    -0.96
     cristales
    -0.94
    ությ
    -0.94
    unächst
    -0.93
     koneksi
    -0.92
    𝐙
    -0.91
     ventanas
    -0.89
    significantly
    -0.89
    POSITIVE LOGITS
     inni
    0.84
    就
    0.82
    ait
    0.80
    <em>
    0.79
     Pensacola
    0.79
    締
    0.79
    í
    0.77
     diable
    0.77
     ۱۳۹
    0.76
    資訊
    0.75
    Activations Density 0.013%

    No Known Activations