© 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-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 23007
    Prev
    Next
    INDEX
    Explanations

    maintaining state or progress

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     двумя
    0.48
    점
    0.46
     слово
    0.45
    銘
    0.45
    艶
    0.44
     classroom
    0.43
    丂
    0.43
     കോ
    0.42
    勘
    0.41
    料金
    0.41
    POSITIVE LOGITS
    isesti
    0.39
    保持
    0.39
     maintain
    0.37
    阻止
    0.37
    waitUntil
    0.36
    ζο
    0.35
     putative
    0.34
    表明
    0.34
     kuni
    0.34
     prevent
    0.33
    Activations Density 0.009%

    No Known Activations