© 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. 39
    Prev
    Next
    INDEX
    Explanations

    numbers and punctuation

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

    The neuron activates on numeric tokens in bibliographic citations—especially years and volume/page numbers.

    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
     også
    -0.92
    cosa
    -0.85
    rugby
    -0.84
    ⁨
    -0.84
     cycling
    -0.82
    缽
    -0.82
     coated
    -0.81
     которому
    -0.79
     sebagainya
    -0.79
     aeropuerto
    -0.77
    POSITIVE LOGITS
     homoto
    0.98
     maaaring
    0.98
     siyang
    0.95
     Là
    0.91
     (!__
    0.87
    نى
    0.86
    ﺔ
    0.86
     admonition
    0.85
    Konto
    0.84
     jäsen
    0.84
    Activations Density 0.001%

    No Known Activations