© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 41274
    Prev
    Next
    INDEX
    Explanations

    advanced

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

    This neuron activates strongly on occurrences of the word “advanced” when it appears as part of technical headings or multi-word titles.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    Helpful
    -1.27
     eiland
    -1.26
     maravillosa
    -1.23
    pecabe
    -1.22
    ApplicationTests
    -1.18
    álna
    -1.17
    Clinic
    -1.16
     executing
    -1.16
    стойчи
    -1.16
     apsau
    -1.16
    POSITIVE LOGITS
     of
    2.02
    Advanced
    1.70
     Advanced
    1.66
     advanced
    1.48
     сейчас
    1.41
     ADVANCED
    1.40
    advanced
    1.38
     avancé
    1.34
     advance
    1.32
    ”?
    1.32
    Activations Density 0.069%

    No Known Activations