© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 114237
    Prev
    Next
    INDEX
    Explanations

    ? or ' or special characters

    np_acts-logits-general · gemini-2.5-flash-lite

    question marks or symbols that appear to be encoding errors or corrupted characters in technical and scientific text.

    oai_token-act-pair · claude-4-5-sonnet
    Triggered by @bfgcn7frf5
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     oddly
    -2.64
     extraordinarily
    -2.58
     undeniably
    -2.52
     often
    -2.50
     unusually
    -2.50
     intensely
    -2.48
     increasing
    -2.41
    In
    -2.41
     immensely
    -2.39
    </h4>
    -2.36
    POSITIVE LOGITS
    งิน
    2.61
     unſ
    2.59
     dando
    2.50
    峮
    2.45
     Dinas
    2.45
    ญี่ป
    2.39
    
    2.38
    ズニー
    2.33
    ymbol
    2.27
    癶
    2.27
    Activations Density 0.029%

    No Known Activations