© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6081
    Prev
    Next
    INDEX
    Explanations

    particle being

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

    The neuron detects passages rich in baseball statistical figures and play‐by‐play numeric details (e.g. innings, at-bats, runs, home runs, pitch counts).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     सपोर्ट
    0.55
     stretcher
    0.52
    run
    0.52
    半島
    0.52
    搶
    0.52
    jaz
    0.52
     নির্যাতন
    0.51
    抢
    0.51
     القيام
    0.50
     yür
    0.50
    POSITIVE LOGITS
     balls
    1.02
    飞
    0.95
     missiles
    0.92
     ball
    0.91
    球
    0.90
     BALL
    0.90
     projectile
    0.88
     projectiles
    0.88
     trajectory
    0.87
     missile
    0.87
    Activations Density 0.250%

    No Known Activations