© 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
    5. 31-GEMMASCOPE-2-RES-262K
    7. 5558
    Prev
    Next
    INDEX
    Explanations

    geometric and logical properties

    np_acts-logits-general · gemini-2.5-flash-lite

    mathematical definitions and properties

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    铁路
    0.68
     финанси
    0.68
    শিয়া
    0.62
    乡村
    0.61
    旄
    0.58
    餐饮
    0.57
     брен
    0.57
    affaires
    0.56
    𝙤
    0.56
     मुकद
    0.55
    POSITIVE LOGITS
     nonzero
    0.93
     trivially
    0.90
     nondegenerate
    0.88
     nontrivial
    0.79
    且
    0.78
     implies
    0.75
     homogeneous
    0.74
     iff
    0.74
     nonnegative
    0.73
     Frobenius
    0.73
    Activations Density 0.029%

    No Known Activations