© 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. 22-GEMMASCOPE-RES-131K
    7. 75174
    Prev
    Next
    INDEX
    Explanations

    events and states described

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

    The neuron strongly responds to words that convey emphasis or noteworthy magnitude—terms like “major,” “huge,” “growth,” “recognition,” and other superlative or high‐impact descriptors.

    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
     it
    -1.20
     sentí
    -1.11
     dejo
    -1.10
     najwięks
    -1.07
     their
    -1.01
    skilled
    -0.99
     miglior
    -0.97
    ially
    -0.96
     vœ
    -0.95
     quiera
    -0.93
    POSITIVE LOGITS
     that
    2.08
    idjan
    1.11
    สำหรับ
    1.02
    issippi
    1.00
    也是
    1.00
    回到
    0.96
    夏天
    0.96
     also
    0.95
     configureStore
    0.95
     indeed
    0.94
    Activations Density 0.012%

    No Known Activations