© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 58276
    Prev
    Next
    INDEX
    Explanations

    japanese "no" particle

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

    The neuron is highly sensitive to the Japanese genitive/particle “no” (as a standalone token in romanized text).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    When
    -2.11
    Because
    -2.00
    During
    -1.92
    when
    -1.77
    Following
    -1.70
    After
    -1.67
    While
    -1.67
    Before
    -1.64
    Despite
    -1.63
    Даже
    -1.57
    POSITIVE LOGITS
     allen
    1.58
     del
    1.45
     —
    1.42
    
    1.41
     coloca
    1.38
     e
    1.37
     toutefois
    1.37
     COST
    1.36
     ending
    1.35
     folosi
    1.35
    Activations Density 0.007%

    No Known Activations