© 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. 55-GEMMASCOPE-2-TRANSCODER-262K
    7. 55071
    Prev
    Next
    INDEX
    Explanations

    early

    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_55_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
     early
    1.18
    early
    1.12
     Early
    1.02
    Early
    0.98
     EARLY
    0.95
    早
    0.95
    早期
    0.84
     temprano
    0.78
     früh
    0.77
     sớm
    0.72
    POSITIVE LOGITS
     حين
    0.39
    瘍
    0.39
     tutup
    0.37
    igste
    0.36
     kiedy
    0.34
     esiste
    0.34
     krijgt
    0.33
     когда
    0.33
     concess
    0.33
    <&
    0.33
    Activations Density 0.003%

    No Known Activations