© 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. 47-GEMMASCOPE-2-TRANSCODER-262K
    7. 218069
    Prev
    Next
    INDEX
    Explanations

    Sun and daylight

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_47_width_262k_l0_small_affine
    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
    /*!
    0.40
    vue
    0.39
    Post
    0.37
     Vue
    0.37
    onial
    0.36
    🇷
    0.36
     singolo
    0.35
     Post
    0.35
     نوف
    0.35
    oner
    0.34
    POSITIVE LOGITS
     plenary
    0.46
     Eddington
    0.45
     জো
    0.43
     leid
    0.43
     doen
    0.40
     Hathaway
    0.39
     Smyth
    0.39
     Wirkung
    0.36
    谠
    0.36
    椃
    0.36
    Activations Density 0.000%

    No Known Activations