© 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. 11095
    Prev
    Next
    INDEX
    Explanations

    Japanese phrases with particles

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

    This neuron fires on Japanese‐language text (i.e. sequences of Japanese characters/words), especially in instructional sentences.

    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
     其他
    1.43
     這個
    1.43
     Ephesians
    1.41
     ବି
    1.40
     waging
    1.39
     prophets
    1.37
     ଅ
    1.36
     时尚
    1.36
     diminishes
    1.35
     ଗ
    1.35
    POSITIVE LOGITS
    な
    1.90
    さ
    1.80
    この
    1.79
    ミ
    1.79
    バ
    1.77
    モ
    1.77
    その
    1.76
    セ
    1.74
    や
    1.73
    い
    1.73
    Activations Density 0.007%

    No Known Activations