© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 213061
    Prev
    Next
    INDEX
    Explanations

    break down

    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_40_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
    Blocking
    0.43
     받고
    0.41
    ِس
    0.39
    ‌است
    0.39
    іга
    0.38
     आक
    0.38
     комна
    0.38
    ifndef
    0.37
     शुभार
    0.37
     도착
    0.37
    POSITIVE LOGITS
     frag
    0.50
    nosed
    0.39
    cheid
    0.39
    frag
    0.39
    喏
    0.38
    付出
    0.37
     القط
    0.37
    strings
    0.37
    string
    0.36
    బాటు
    0.36
    Activations Density 0.000%

    No Known Activations