© 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. 53984
    Prev
    Next
    INDEX
    Explanations

    scored goals or points

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

    The neuron strongly activates on numeric tokens (scores, stats, weights, and other figures) in sports reporting.

    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
    だけでなく
    -0.91
     whoſe
    -0.87
     konsek
    -0.87
     konkret
    -0.85
     weiterer
    -0.84
     lande
    -0.84
     weiteren
    -0.83
    chés
    -0.82
    tola
    -0.81
     `${
    -0.81
    POSITIVE LOGITS
     but
    1.26
     only
    1.12
     tries
    1.09
    Hanya
    1.00
     intentar
    0.99
     spät
    0.99
     nekaj
    0.98
     attempt
    0.96
     try
    0.95
     attempting
    0.95
    Activations Density 0.011%

    No Known Activations