© 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. 10847
    Prev
    Next
    INDEX
    Explanations

    change

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

    This neuron is triggered by formal domain-specific terminology—specialized content-heavy nouns and phrases from policy, legal, technical, or research contexts.

    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
    絶妙
    -0.90
    מוס
    -0.87
    magini
    -0.81
    SUGGEST
    -0.81
    あえて
    -0.80
     Necessary
    -0.79
    loginPage
    -0.79
     ジップ
    -0.79
    呼ぶ
    -0.78
     isola
    -0.76
    POSITIVE LOGITS
     changes
    1.17
     changed
    1.11
     original
    1.10
     change
    1.08
     changing
    1.02
    Changing
    0.98
     menangani
    0.98
    Change
    0.96
     altering
    0.96
     mengubah
    0.93
    Activations Density 0.032%

    No Known Activations