© 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. 5579
    Prev
    Next
    INDEX
    Explanations

    1990s

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

    This neuron strongly activates on numeric tokens (especially multi-digit numbers and numeric measurements).

    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
     १८४
    0.70
    ⊱
    0.69
    कार
    0.68
     하였다
    0.67
     இருந்து
    0.65
    १८४
    0.65
    ба
    0.64
    ните
    0.63
     idő
    0.63
     כמו
    0.63
    POSITIVE LOGITS
    9
    1.13
    ９
    1.10
     nineties
    1.01
    ៩
    0.94
    ९
    0.91
     ninety
    0.87
    ၉
    0.86
    ৯
    0.84
    ༩
    0.81
    ۹
    0.79
    Activations Density 0.191%

    No Known Activations