© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 5864
    Prev
    Next
    INDEX
    Explanations

    initial temperatures

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

    This neuron strongly activates for the standalone token “Danish,” i.e. when the Danish language is mentioned.

    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
     y
    0.79
     כ
    0.78
     Iglesia
    0.78
     ون
    0.75
     DEI
    0.75
    aas
    0.75
    y
    0.75
     ย
    0.74
    หมือน
    0.74
     ی
    0.73
    POSITIVE LOGITS
    我们
    0.84
    著作
    0.80
    처
    0.80
    人民政府
    0.79
    婶
    0.79
    我们在
    0.78
    叭
    0.78
    提高
    0.77
     lessen
    0.77
    只限
    0.77
    Activations Density 0.000%

    No Known Activations