© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 106508
    Prev
    Next
    INDEX
    Explanations

    Japanese language tokens and following words

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

    The neuron detects spans of East Asian (CJK) text, i.e. sequences of Chinese or Japanese characters.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     an
    -2.25
     only
    -2.13
     because
    -2.05
     before
    -2.05
     just
    -2.03
     when
    -2.00
     with
    -1.98
     after
    -1.97
     for
    -1.96
     at
    -1.96
    POSITIVE LOGITS
     Dinas
    1.78
    primarily
    1.74
    creating
    1.69
    컹
    1.66
     faciliter
    1.63
    possibly
    1.61
    several
    1.59
    튿
    1.59
    whose
    1.59
    commonly
    1.58
    Activations Density 0.012%

    No Known Activations