© 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. 53845
    Prev
    Next
    INDEX
    Explanations

    Should followed by pronoun or noun

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

    starts with should

    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
    ść
    0.74
     丿
    0.58
     sposób
    0.58
     усі
    0.57
     jeżeli
    0.56
     усіх
    0.56
    每一个
    0.56
     bewust
    0.55
     قابل
    0.54
    為什麼
    0.54
    POSITIVE LOGITS
    ל
    0.72
    le
    0.70
    al
    0.66
    J
    0.63
    en
    0.62
     }{
    0.60
     onerror
    0.60
    centi
    0.56
    g
    0.56
    W
    0.56
    Activations Density 0.035%

    No Known Activations