© 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-262K
    7. 136449
    Prev
    Next
    INDEX
    Explanations

    creating directories with -p

    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
    面上
    0.41
    未
    0.40
    分かり
    0.39
     未
    0.39
     वर्च
    0.38
     القسم
    0.37
    बाला
    0.37
     スマート
    0.37
    lio
    0.37
     Pacquiao
    0.36
    POSITIVE LOGITS
     workings
    0.42
     prefixed
    0.38
     millet
    0.38
    㯎
    0.37
     nucleon
    0.37
    ennies
    0.35
    igheden
    0.35
    परेटर
    0.34
    스키
    0.34
    ето
    0.34
    Activations Density 0.000%

    No Known Activations