© 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. 28477
    Prev
    Next
    INDEX
    Explanations

    called Apostles

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

    The neuron fires on numeric tokens (integers, decimals, and related symbols like percentages) in the 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
    дру
    -0.95
    ENDIAN
    -0.91
     Halloween
    -0.90
    Gemeinden
    -0.88
    SUPPORTED
    -0.88
     penis
    -0.86
    älä
    -0.86
     Proposition
    -0.83
    ТУ
    -0.81
    assertSame
    -0.81
    POSITIVE LOGITS
     registrazione
    0.98
     cantit
    0.87
     facendo
    0.86
     sezione
    0.85
     herunterladen
    0.84
     barbaric
    0.84
     presenza
    0.82
     partea
    0.81
     avendo
    0.81
     estancia
    0.80
    Activations Density 0.042%

    No Known Activations