© 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. 10-GEMMASCOPE-RES-131K
    7. 119524
    Prev
    Next
    INDEX
    Explanations

    specific entities

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

    The neuron fires on foreign‐origin terms—especially loanwords and proper names from various languages (e.g. Spanish, Italian, Japanese, Indian) rather than native English 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_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
    霂
    -2.28
     цього
    -2.14
     poursuite
    -1.97
    裃
    -1.93
    蹓
    -1.92
     diffic
    -1.88
     responsabilité
    -1.88
    馘
    -1.88
     réalisation
    -1.87
    渃
    -1.87
    POSITIVE LOGITS
     Our
    2.31
     notre
    2.30
     We
    2.03
     is
    2.02
     Are
    1.93
    ѣмъ
    1.91
     l
    1.85
     в
    1.84
    鹛
    1.78
     nuestra
    1.77
    Activations Density 0.005%

    No Known Activations