© 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. 40-GEMMASCOPE-2-RES-262K
    7. 2688
    Prev
    Next
    INDEX
    Explanations

    error handling in code

    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
    '>";
    0.40
    滟
    0.40
    gramModel
    0.39
    营造
    0.38
     terraz
    0.38
    OMBRE
    0.38
    ówno
    0.38
    🚱
    0.37
    Kul
    0.37
    Muito
    0.37
    POSITIVE LOGITS
     Pair
    0.41
    ht
    0.35
     Steering
    0.35
     contributes
    0.34
     Airline
    0.34
     })
    0.33
     Insurance
    0.33
     would
    0.33
     President
    0.33
     EL
    0.33
    Activations Density 0.050%

    No Known Activations