© 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. 26824
    Prev
    Next
    INDEX
    Explanations

    dimensions and measurements

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

    The neuron detects numeric dimension specifications (measurements given as lengths × widths × heights, often with units and quotation marks).

    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
     even
    -1.04
     Polish
    -0.82
     as
    -0.80
     mellom
    -0.79
    ％
    -0.77
    ารถ
    -0.75
    🧣
    -0.74
    లా
    -0.73
     between
    -0.72
     Even
    -0.72
    POSITIVE LOGITS
     assass
    0.92
    দ্ধ
    0.86
     ieder
    0.84
    wechs
    0.84
    ilirsiniz
    0.84
    жный
    0.83
     pourrait
    0.82
    Blanc
    0.79
     trám
    0.79
    levé
    0.79
    Activations Density 0.039%

    No Known Activations