© 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. 18751
    Prev
    Next
    INDEX
    Explanations

    decrease or subtraction

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

    reduce, minimize, decrease

    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
     нара
    0.46
    無理
    0.44
    accro
    0.42
     ***!
    0.42
    糍
    0.42
     ৷
    0.40
    _{+}$
    0.40
    それでも
    0.39
    oczy
    0.39
    顽
    0.39
    POSITIVE LOGITS
     subtract
    0.92
     subtracted
    0.90
    Subtract
    0.86
     Subtract
    0.85
     subtracting
    0.84
     subtraction
    0.83
     decreases
    0.82
     decrease
    0.79
     decreased
    0.79
     detract
    0.77
    Activations Density 0.076%

    No Known Activations