© 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. 24-GEMMASCOPE-2-RES-16K
    7. 13562
    Prev
    Next
    INDEX
    Explanations

    "is a" pattern

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

    The neuron fires strongly on auxiliary‐verb sequences—especially “does,” “have to,” and “be” phrases (e.g. “it doesn’t have to be,” “does it work,” “turned out to be”).

    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_24_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
     ఉన్నారు
    0.91
     ønsker
    0.79
     רבים
    0.79
    ต้องการ
    0.74
     जानता
    0.73
    넋
    0.72
    ຈັດສົ່ງ
    0.70
    อยาก
    0.70
     έχ
    0.69
    ייתה
    0.69
    POSITIVE LOGITS
     raining
    0.89
     snowing
    0.87
     pertains
    0.74
     Happened
    0.72
     Comme
    0.71
     transpired
    0.71
     comme
    0.68
    g
    0.67
     dawned
    0.67
     that
    0.66
    Activations Density 0.999%

    No Known Activations