© 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. 34-GEMMASCOPE-RES-131K
    7. 81509
    Prev
    Next
    INDEX
    Explanations

    cricket runs

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

    The neuron activates on numeric tokens—i.e. digits or numbers representing counts, scores, overs, etc.

    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
     verschillen
    -0.85
     じ
    -0.85
    oked
    -0.83
     Iliad
    -0.81
    arum
    -0.81
    ⟷
    -0.80
     QP
    -0.79
     intenciones
    -0.78
    TW
    -0.77
    င
    -0.75
    POSITIVE LOGITS
     off
    1.13
     in
    0.85
    饅
    0.82
    取材
    0.79
    рь
    0.79
    acakt
    0.76
     balls
    0.76
    пуск
    0.75
    ох
    0.75
     pointer
    0.75
    Activations Density 0.009%

    No Known Activations