© 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. 136458
    Prev
    Next
    INDEX
    Explanations

    Explaining performance analysis

    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
     பகி
    0.43
     leveraging
    0.41
    っていう
    0.40
     seamlessly
    0.40
    👉
    0.39
     linking
    0.39
     connecting
    0.38
     یکی
    0.38
    engkap
    0.38
    linking
    0.38
    POSITIVE LOGITS
    Ranges
    0.41
     sayıda
    0.39
    putable
    0.38
    🥃
    0.38
    ranged
    0.37
     уту
    0.37
     workmen
    0.37
     workers
    0.36
    sj
    0.36
    工人
    0.36
    Activations Density 0.006%

    No Known Activations