© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 71815
    Prev
    Next
    INDEX
    Explanations

    extraordinary or spiritual qualities

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

    The neuron fires on highly promotional, over-the-top adjectives or nouns (e.g. “magical,” “super,” “celestial,” “biblical,” “mystical,” “wonderful”) used to hype or glorify a subject.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     magnific
    -1.33
    泚
    -1.32
    膩
    -1.29
     více
    -1.26
     víc
    -1.25
    categorized
    -1.23
    cette
    -1.23
     мой
    -1.20
     SEVER
    -1.20
     turmeric
    -1.20
    POSITIVE LOGITS
     that
    1.65
     truly
    1.56
     only
    1.51
     because
    1.47
     when
    1.43
     within
    1.42
     people
    1.38
     there
    1.34
     of
    1.32
     really
    1.30
    Activations Density 0.038%

    No Known Activations