© 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. 22-GEMMASCOPE-RES-131K
    7. 73253
    Prev
    Next
    INDEX
    Explanations

    overall statistics and final standings

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

    The neuron activates on numeric data tokens—i.e. statistical or table values (years, scores, counts) 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
     several
    -0.84
    ellum
    -0.74
     allerdings
    -0.74
    径
    -0.73
     ruedas
    -0.73
     условиях
    -0.73
    owano
    -0.73
    ayanan
    -0.73
    関連リンク
    -0.72
     fenêtres
    -0.72
    POSITIVE LOGITS
     overall
    2.06
    overall
    1.66
    Overall
    1.49
     insgesamt
    1.38
     Overall
    1.38
     totaled
    1.28
     cumulative
    1.27
     totalled
    1.22
     statistics
    1.21
     averaged
    1.20
    Activations Density 0.044%

    No Known Activations