© 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. 54315
    Prev
    Next
    INDEX
    Explanations

    common pairings after words

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

    This neuron fires on relatively rare or content-heavy tokens—i.e. uncommon, domain-specific, or low-frequency words.

    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
     開始
    -1.23
     незавер
    -1.07
     封面
    -1.04
     arou
    -1.01
     그것
    -0.99
    ถ้า
    -0.98
    選ぶ
    -0.95
    ｼ
    -0.94
    　　《
    -0.94
    每個
    -0.94
    POSITIVE LOGITS
     devoir
    1.12
     ngunit
    1.11
    以外の
    1.11
     vendedores
    1.08
    水が
    1.05
    wiek
    1.05
     dimentic
    1.03
     pintado
    1.02
     Euró
    1.02
    の中の
    1.01
    Activations Density 0.104%

    No Known Activations