© 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. 71229
    Prev
    Next
    INDEX
    Explanations

    sports event stages and results

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

    The neuron is primarily activated by numeric tokens—years, scores, draw sizes, seed numbers, and other figures.

    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
    jointe
    -1.20
     OBITUARY
    -1.10
    ﹕
    -0.99
     flott
    -0.96
     beiden
    -0.96
     filmen
    -0.95
    :「
    -0.94
    application
    -0.94
     APPLICATION
    -0.93
    </h1>
    -0.92
    POSITIVE LOGITS
    espère
    1.24
    mények
    0.99
    ustes
    0.96
    НИМА
    0.91
    andidaten
    0.91
    ktive
    0.91
     offenbar
    0.91
     alguno
    0.90
     culoare
    0.88
    或者是
    0.88
    Activations Density 0.004%

    No Known Activations