© 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-3-4B-IT
    5. 29-GEMMASCOPE-2-RES-262K
    7. 145352
    Prev
    Next
    INDEX
    Explanations

    contrastive conjunctions and clauses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/resid_post/layer_29_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    我在
    0.84
     eller
    0.73
     in
    0.73
    கர்
    0.71
    你在
    0.70
     -
    0.68
     bluetooth
    0.67
     Requested
    0.67
     "
    0.67
     або
    0.67
    POSITIVE LOGITS
     nhưng
    1.31
     mientras
    1.30
     despite
    1.29
     while
    1.26
     لیکن
    1.26
     tetapi
    1.25
     pero
    1.19
     किन्तु
    1.17
     लेकिन
    1.17
     apesar
    1.17
    Activations Density 0.001%

    No Known Activations