© 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. 31-GEMMASCOPE-2-RES-262K
    7. 7876
    Prev
    Next
    INDEX
    Explanations

    acknowledging imperfection and inevitability

    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_31_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
    द्वारे
    0.37
    user
    0.36
    encoding
    0.36
     firebase
    0.33
     toj
    0.33
    apache
    0.33
    {\$
    0.33
    ören
    0.32
     apache
    0.32
    簍
    0.32
    POSITIVE LOGITS
     sometimes
    0.71
     inevitable
    0.69
     inev
    0.65
     parfois
    0.64
     imperfect
    0.64
     inevitably
    0.64
     Sometimes
    0.63
    Sometimes
    0.63
     unavoid
    0.63
     иногда
    0.62
    Activations Density 0.294%

    No Known Activations