© 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. 37-GEMMASCOPE-2-TRANSCODER-262K
    7. 201262
    Prev
    Next
    INDEX
    Explanations

    error handling or exceptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_37_width_262k_l0_small_affine
    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
     ވަ
    0.44
    چھے
    0.43
    rupulous
    0.41
    <start_of_turn>
    0.39
     pact
    0.39
     시스템
    0.39
     perjanjian
    0.38
    制度
    0.38
     Yayın
    0.37
    獲得
    0.37
    POSITIVE LOGITS
    Ignore
    0.42
    ignore
    0.38
     exceptions
    0.38
    寞
    0.36
     ignore
    0.36
     exception
    0.35
    hint
    0.35
     ignoring
    0.35
    Diabetes
    0.35
     prefix
    0.35
    Activations Density 0.003%

    No Known Activations