© 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. 1551
    Prev
    Next
    INDEX
    Explanations

    algorithmic complexity and performance

    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
     figurines
    0.43
    userManager
    0.40
    規範
    0.40
    衞
    0.39
     consentimiento
    0.39
    ইসলাম
    0.39
     sociedades
    0.39
     কমন
    0.39
    竣工
    0.38
    鸷
    0.38
    POSITIVE LOGITS
     inefficient
    1.01
     worst
    0.95
     inefficiencies
    0.89
     inefficiency
    0.89
    Worst
    0.86
    worst
    0.86
     Worst
    0.86
     computationally
    0.84
     asymptotically
    0.83
     efficiency
    0.82
    Activations Density 0.075%

    No Known Activations