© 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. 34-GEMMASCOPE-RES-131K
    7. 31364
    Prev
    Next
    INDEX
    Explanations

    honored and selected people

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

    The neuron specifically lights up on numeric tokens (e.g. years, figures, percentages).

    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
     for
    -1.14
     Even
    -1.05
     when
    -1.02
     as
    -0.99
     even
    -0.95
     three
    -0.94
     so
    -0.93
    Even
    -0.91
     or
    -0.91
     while
    -0.90
    POSITIVE LOGITS
     nomine
    1.08
    歺
    1.08
     sélectionner
    1.05
     nomi
    1.04
     SECRET
    1.03
     nomina
    1.02
     Filosof
    1.00
     maniere
    1.00
     nobili
    0.99
     notori
    0.98
    Activations Density 0.050%

    No Known Activations