© 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. 34-GEMMASCOPE-RES-131K
    7. 59714
    Prev
    Next
    INDEX
    Explanations

    articles starting with questions, titles, or proper nouns

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

    The neuron fires on numeric literal tokens (especially multi‐digit numbers) in code or comments.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Արտ
    -0.92
    зирован
    -0.83
    While
    -0.83
     adulta
    -0.82
     piloto
    -0.81
     навіть
    -0.79
    Does
    -0.77
     compacto
    -0.77
     even
    -0.77
     Autres
    -0.77
    POSITIVE LOGITS
    нового
    0.93
     admirably
    0.89
    Université
    0.83
    ជ
    0.82
     Analyse
    0.82
    EndInit
    0.81
    жте
    0.80
     interloc
    0.80
     iceberg
    0.80
    mond
    0.79
    Activations Density 0.000%

    No Known Activations