© 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. 135444
    Prev
    Next
    INDEX
    Explanations

    safety guards and bits

    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
     убы
    0.39
    ీయ
    0.38
    rać
    0.38
    发电
    0.38
    vertine
    0.37
    لیت
    0.37
     প্রজ
    0.37
    甶
    0.36
    Santos
    0.36
    र्घ
    0.35
    POSITIVE LOGITS
     Guards
    0.50
     Guard
    0.46
     guard
    0.44
     bit
    0.43
     guards
    0.43
     adjustment
    0.42
    bit
    0.42
     measuring
    0.42
     unintentional
    0.41
    guard
    0.41
    Activations Density 0.005%

    No Known Activations