© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 34561
    Prev
    Next
    INDEX
    Explanations

    wherever possible

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

    The neuron specializes in spotting quantitative or measurement-oriented terms—numbers, costs/savings language, statistical or metric words (e.g. “saving,” “money,” “expenses,” “coordinates,” “tests,” “evaluation”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    seguida
    -1.02
    ;%%
    -1.02
    点が
    -0.99
    以外にも
    -0.99
     impide
    -0.98
    IMENTO
    -0.95
    лька
    -0.94
    ٥
    -0.94
    RAEL
    -0.94
     onu
    -0.93
    POSITIVE LOGITS
     wherever
    5.53
     where
    4.88
     whenever
    4.69
    Wherever
    3.69
     Wherever
    3.67
    whenever
    3.50
    where
    3.27
    Where
    3.16
     Whenever
    3.13
     if
    2.89
    Activations Density 0.147%

    No Known Activations