© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 28316
    Prev
    Next
    INDEX
    Explanations

    filtering and normalization

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

    The neuron flags numeric tokens—literal numbers or sequences of digits—in code or text.

    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
    entrion
    -0.87
     newBuilder
    -0.87
     merchants
    -0.85
     neurons
    -0.85
    heten
    -0.82
    ⏹
    -0.82
     pregunta
    -0.81
     autres
    -0.81
     ozone
    -0.81
    ItemGroup
    -0.79
    POSITIVE LOGITS
     final
    1.03
    ruka
    0.89
    ∏
    0.87
    もう一度
    0.87
    licante
    0.85
    final
    0.84
    eX
    0.84
    cuss
    0.83
     muhimu
    0.81
     Final
    0.81
    Activations Density 0.011%

    No Known Activations