© 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. 53-GEMMASCOPE-2-RES-262K
    7. 38325
    Prev
    Next
    INDEX
    Explanations

    adding buffer or margin

    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_53_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
     harmful
    0.44
    કા
    0.39
     intensely
    0.38
     harming
    0.38
    噻
    0.36
     dépl
    0.36
     interrump
    0.36
     harmed
    0.36
    強く
    0.36
    🕍
    0.36
    POSITIVE LOGITS
     buffer
    1.80
     margin
    1.79
     margins
    1.66
     buffers
    1.60
    buffer
    1.56
    Buffer
    1.54
    margin
    1.54
     Margin
    1.53
     Buffer
    1.51
     margen
    1.49
    Activations Density 0.031%

    No Known Activations