© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-65K
    7. 51059
    Prev
    Next
    INDEX
    Explanations

    favor composition over inheritance

    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_65k_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
    getStock
    0.40
     सईद
    0.39
     조사
    0.38
    🏾
    0.38
    🍶
    0.38
    eston
    0.37
    oton
    0.37
    estock
    0.37
    STOCK
    0.37
    nić
    0.37
    POSITIVE LOGITS
    Composition
    0.51
     Composition
    0.50
     компози
    0.50
     composition
    0.49
     Preferences
    0.44
    composition
    0.43
    compos
    0.43
     предпоч
    0.43
    組
    0.42
     compositions
    0.42
    Activations Density 0.010%

    No Known Activations