© 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. 41602
    Prev
    Next
    INDEX
    Explanations

    German verbs followed by modal verbs

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

    können and related verbs

    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
     στον
    0.53
    ח
    0.50
    していますが
    0.47
    そして
    0.47
     ಮತ್ತು
    0.47
     във
    0.47
    とりあえず
    0.47
    在
    0.46
    とお
    0.46
     ਅਤੇ
    0.46
    POSITIVE LOGITS
     könne
    0.53
     können
    0.49
     konnten
    0.48
     möchte
    0.47
     mussten
    0.47
     könnten
    0.47
     either
    0.46
     Claire
    0.46
     wollten
    0.46
     sollen
    0.45
    Activations Density 0.002%

    No Known Activations