© 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. 55595
    Prev
    Next
    INDEX
    Explanations

    Initial checks and steps

    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
    म
    0.46
    드
    0.45
    不
    0.44
     ри
    0.43
    ד
    0.43
     소
    0.42
     بر
    0.41
    г
    0.41
     כ
    0.41
    ガス
    0.40
    POSITIVE LOGITS
     aaye
    0.45
     tqdm
    0.44
     kanë
    0.43
     stunned
    0.43
    ಿದ್ದಾನೆ
    0.43
     lLoginID
    0.41
     vagu
    0.40
     আসছেন
    0.40
    ಿದ್ದಾರೆ
    0.39
    `],
    0.39
    Activations Density 0.001%

    No Known Activations