© 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. 1521
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on capitalized names and technical acronyms (e.g. brand names, file‐format or API abbreviations).

    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
    (
    1.10
    （
    1.04
     (
    0.89
    (!
    0.86
    ()
    0.85
    y
    0.82
    (,
    0.81
    ((
    0.79
    l
    0.79
     $($
    0.79
    POSITIVE LOGITS
    
    0.95
     ktorí
    0.83
     którzy
    0.79
     które
    0.78
     kteří
    0.77
     แห่ง
    0.77
    ವನ್ನು
    0.77
     которая
    0.71
    그리고
    0.71
     açısından
    0.71
    Activations Density 0.632%

    No Known Activations