© 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. 22-GEMMASCOPE-2-TRANSCODER-262K
    7. 45270
    Prev
    Next
    INDEX
    Explanations

    repetitive tasks or 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/transcoder_all/layer_22_width_262k_l0_small_affine
    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
     कीमतें
    0.35
    燴
    0.34
    後的
    0.33
    álisis
    0.32
     seize
    0.32
    एनएल
    0.32
    azor
    0.31
    
    0.31
    sizing
    0.31
     seizing
    0.31
    POSITIVE LOGITS
     unnecessary
    0.41
     अनावश्यक
    0.38
    Waste
    0.38
     waste
    0.36
    浪费
    0.35
    redund
    0.34
     Waste
    0.34
    waste
    0.34
    Progress
    0.33
    Preliminary
    0.33
    Activations Density 0.077%

    No Known Activations