© 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-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-16K
    7. 12626
    Prev
    Next
    INDEX
    Explanations

    record scratch, book, keeping, label

    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_16k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    ோ
    0.46
     রাশিয়া
    0.46
     "__
    0.41
    ">—
    0.41
     executed
    0.41
    াস
    0.40
     corpse
    0.40
    )²
    0.40
    ிறார்கள்
    0.40
     মর্যা
    0.40
    POSITIVE LOGITS
    Record
    0.71
     Record
    0.68
    keeping
    0.63
    レコード
    0.56
    C
    0.56
    KEEP
    0.55
    record
    0.54
    breaking
    0.53
    RECORD
    0.53
    Keep
    0.52
    Activations Density 0.037%

    No Known Activations