© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 40-GEMMASCOPE-2-RES-16K
    7. 10274
    Prev
    Next
    INDEX
    Explanations

    tacky execution

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

    cliche, boring, or vague statements

    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
     speculate
    0.48
     mischief
    0.45
     mischievous
    0.45
     speculating
    0.45
     imperfect
    0.44
     tumultuous
    0.44
    乐观
    0.44
    噂
    0.44
     wicked
    0.43
     heartbreaking
    0.43
    POSITIVE LOGITS
     boring
    1.72
     Boring
    1.46
     monotonous
    1.41
     monotony
    1.37
     bored
    1.34
     boredom
    1.23
     bland
    1.20
     monoton
    1.20
     bores
    1.20
     Bored
    1.17
    Activations Density 0.428%

    No Known Activations