© 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. 25793
    Prev
    Next
    INDEX
    Explanations

    code punctuation and delimiters

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

    This neuron detects import‐statement syntax (especially the “from” keyword, module‐path tokens, and the trailing semicolon in import 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_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
    щие
    -0.86
    izioni
    -0.86
     Você
    -0.81
    
    -0.80
    klá
    -0.79
     vomit
    -0.77
     كذلك
    -0.77
    vited
    -0.77
    Msgs
    -0.77
    oree
    -0.76
    POSITIVE LOGITS
    ánu
    0.84
    übt
    0.84
    uldu
    0.83
     Planets
    0.83
     (*.
    0.82
     детали
    0.82
     порта
    0.82
     Orbital
    0.81
    /−
    0.81
     lampa
    0.81
    Activations Density 0.016%

    No Known Activations