© 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. 22-GEMMASCOPE-RES-131K
    4. 68971
    Prev
    Next
    INDEX
    Explanations

    advantage, suicide, Independence, excuse, tests, victory, job, tip, way, means

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

    This neuron activates on nouns (including noun heads and their preceding articles/determiners).

    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
    そこで
    -0.96
     clayey
    -0.88
    egl
    -0.85
     derive
    -0.84
     may
    -0.83
    specifier
    -0.83
    ksjon
    -0.81
    notifier
    -0.80
    AU
    -0.78
    intellij
    -0.78
    POSITIVE LOGITS
    manteau
    1.01
    tothe
    0.99
     helpless
    0.96
     fuis
    0.96
    romeda
    0.94
     kiti
    0.93
     respek
    0.93
     garan
    0.93
     COOK
    0.92
     scuola
    0.92
    Activations Density 0.060%

    No Known Activations