© 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-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 1871
    Prev
    Next
    INDEX
    Explanations

    Korean and Japanese grammatical endings

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

    こと (koto) / likely verb endings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_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
     दिलेल्या
    0.58
    पिछले
    0.57
    ي
    0.56
     झालेल्या
    0.55
    isierten
    0.54
     verilen
    0.53
    கடந்த
    0.53
     কর্তৃক
    0.52
    áneo
    0.52
     सकाळी
    0.52
    POSITIVE LOGITS
     것이
    0.88
     것
    0.82
    ことを
    0.79
    ように
    0.77
     것은
    0.76
     것처럼
    0.76
    ような
    0.74
     것으로
    0.72
     곳
    0.72
    のが
    0.71
    Activations Density 0.005%

    No Known Activations