© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 10196
    Prev
    Next
    INDEX
    Explanations

    stability and power

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

    The main thing this neuron does is detect numeric tokens, especially sports‐related statistics and 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
     inaccur
    0.54
     फेंक
    0.52
     sutures
    0.52
     αντα
    0.52
    抛
    0.52
    Accuracy
    0.49
     حفظ
    0.47
    戍
    0.46
     throws
    0.46
    ochond
    0.46
    POSITIVE LOGITS
     batting
    1.00
     batter
    0.92
     batters
    0.92
     bat
    0.90
     Bat
    0.87
    bat
    0.86
     hitter
    0.86
     batsman
    0.86
     Batter
    0.85
     batsmen
    0.84
    Activations Density 0.123%

    No Known Activations