© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 110783
    Prev
    Next
    INDEX
    Explanations

    ugly, bland, boring descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mischie
    -1.01
     slower
    -0.93
    묘
    -0.90
     properly
    -0.89
     stanzas
    -0.88
    いれば
    -0.88
    ্ষ
    -0.86
    細い
    -0.85
    ちょっとした
    -0.84
    起き
    -0.82
    POSITIVE LOGITS
     ugly
    1.41
    ugly
    1.22
     bulky
    1.20
     bland
    1.18
     generic
    1.11
    generic
    1.05
     unattractive
    1.04
     utilitarian
    1.03
    boring
    1.02
     boring
    1.00
    Activations Density 0.057%

    No Known Activations