© 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. 28400
    Prev
    Next
    INDEX
    Explanations

    technical terms

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

    The neuron strongly activates on technical acronyms and specialized scientific terms, particularly all-caps abbreviations (e.g. SEM, TNF, PL, ND) and similar domain-specific tokens.

    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
    ,
    -1.02
    стоя
    -0.91
     atrop
    -0.90
     ecológica
    -0.90
    <em>
    -0.89
    ulaires
    -0.85
     mete
    -0.85
     шоу
    -0.84
     мин
    -0.83
    ırım
    -0.82
    POSITIVE LOGITS
     dirancang
    0.97
    scaleY
    0.96
    ֶּ
    0.96
     balcón
    0.95
     novelas
    0.93
    쩡
    0.93
    τέ
    0.91
    ներ
    0.91
     edição
    0.90
    LARE
    0.90
    Activations Density 0.219%

    No Known Activations