© 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. 35837
    Prev
    Next
    INDEX
    Explanations

    foreign language words

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

    This neuron responds to the impersonal, formulaic phrasing typical of formal technical or patent‐style exposition (e.g. “it is common to…,” “involves the…,” “using a…”).

    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
     because
    -1.10
     since
    -1.02
     навіть
    -1.01
     лікування
    -0.99
    ၇
    -0.97
     rodin
    -0.96
     тільки
    -0.93
     Erster
    -0.93
     немає
    -0.93
    리고
    -0.91
    POSITIVE LOGITS
    运
    1.01
    ごめん
    1.00
     Maßnahmen
    1.00
    Conoce
    0.98
     怎样
    0.97
    为
    0.96
    选择
    0.94
     consultez
    0.93
    bilidad
    0.91
    Ine
    0.91
    Activations Density 0.623%

    No Known Activations