© 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. 50-GEMMASCOPE-2-TRANSCODER-262K
    7. 201158
    Prev
    Next
    INDEX
    Explanations

    pellets

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_50_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    Basel
    0.38
    ┕
    0.37
     وز
    0.36
     Basel
    0.36
     Freestyle
    0.36
    冢
    0.36
     Zurich
    0.35
    派
    0.35
    ฤษ
    0.35
    梗
    0.34
    POSITIVE LOGITS
    Pel
    3.22
     Pel
    3.16
     pel
    3.08
    pel
    2.92
     PEL
    2.72
    PEL
    2.67
     Pell
    2.52
     pelvic
    2.50
     pellet
    2.48
     pellets
    2.42
    Activations Density 0.004%

    No Known Activations