© 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. 40-GEMMASCOPE-2-RES-16K
    7. 8763
    Prev
    Next
    INDEX
    Explanations

    magic, witches, and magicians

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

    magic and magic users

    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_40_width_16k_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
    oxygen
    0.47
    भाल
    0.45
     oxygen
    0.44
    蛋白質
    0.43
    욘
    0.42
    chromosome
    0.42
     patriots
    0.42
     chutes
    0.42
     ऑक्सीजन
    0.41
     oxígeno
    0.41
    POSITIVE LOGITS
     witchcraft
    1.70
     magick
    1.68
     occult
    1.58
     witches
    1.58
    魔法
    1.52
     witch
    1.49
     magicians
    1.48
     Witch
    1.45
     magician
    1.45
    🧙
    1.43
    Activations Density 0.337%

    No Known Activations