© 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. 31-GEMMASCOPE-2-RES-16K
    7. 10719
    Prev
    Next
    INDEX
    Explanations

    undermining domestic happiness

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

    The neuron fires whenever it sees very high-frequency function words—especially the definite article “the” (often together with words like “that,” “for,” “all,” etc.)

    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
    보다는
    0.65
    දි
    0.59
     പ്രശ്
    0.58
    each
    0.57
     مختلف
    0.57
     वेगवेगळ्या
    0.57
     ／
    0.57
    略
    0.55
     مختلفة
    0.54
    漏
    0.54
    POSITIVE LOGITS
    本来
    1.26
    原本
    1.22
     cherished
    1.15
     treasured
    1.09
     originally
    1.04
     beloved
    1.03
     normally
    0.98
     hitherto
    0.94
     originalmente
    0.94
     innocent
    0.94
    Activations Density 0.602%

    No Known Activations