© 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-262K
    7. 215268
    Prev
    Next
    INDEX
    Explanations

    design for durability performance scalability

    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_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     próxim
    0.46
     জন্যে
    0.46
     consideration
    0.45
    avak
    0.45
     rozpoc
    0.43
     commencent
    0.43
     kasutada
    0.43
     justifications
    0.43
     koristiti
    0.41
     commen
    0.40
    POSITIVE LOGITS
     stability
    0.63
     success
    0.63
     accuracy
    0.59
    stability
    0.58
     эффективность
    0.57
     snelheid
    0.56
     correctness
    0.55
     скорость
    0.55
    success
    0.55
     speed
    0.53
    Activations Density 0.025%

    No Known Activations