© 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. 2147
    Prev
    Next
    INDEX
    Explanations

    pronunciation guides in parentheses

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

    This neuron activates strongly on the Japanese katakana term “アンダーフィル剤” (underfill agent).

    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
     հ
    0.75
     الن
    0.75
    اح
    0.74
     تد
    0.73
     اله
    0.72
    جع
    0.72
     Δια
    0.71
     ق
    0.70
    خ
    0.70
     نو
    0.69
    POSITIVE LOGITS
    －
    1.21
    ＝
    1.05
    ／
    1.00
    ・
    0.97
    ＆
    0.96
    ＋
    0.95
    즈
    0.94
    ング
    0.91
    ‧
    0.90
    －－
    0.88
    Activations Density 0.034%

    No Known Activations