© 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. 215137
    Prev
    Next
    INDEX
    Explanations

    design

    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
    अधिका
    0.44
     unequiv
    0.41
    రికీ
    0.41
     Тихо
    0.40
     relocating
    0.39
     நாடு
    0.39
     perinatal
    0.39
    𒋗
    0.39
     unequivocally
    0.39
    пах
    0.38
    POSITIVE LOGITS
     mất
    0.41
     output
    0.40
     transaction
    0.38
     Output
    0.37
    BAR
    0.36
    Scan
    0.36
     swe
    0.36
    Transaction
    0.36
    Output
    0.36
    ².
    0.35
    Activations Density 0.001%

    No Known Activations