© 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. 41776
    Prev
    Next
    INDEX
    Explanations

    infinitely repeating patterns

    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
    :
    0.56
     are
    0.53
    ::
    0.53
    ;
    0.52
     and
    0.52
    '
    0.52
       
    0.51
    us
    0.51
    )
    0.51
          
    0.48
    POSITIVE LOGITS
     کہانی
    0.66
     करियर
    0.63
     showbiz
    0.61
     चर्चित
    0.59
     stardom
    0.58
     lackluster
    0.57
    Noeud
    0.57
     داستان
    0.55
     कहानी
    0.55
     lovable
    0.55
    Activations Density 0.003%

    No Known Activations