© 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. 48258
    Prev
    Next
    INDEX
    Explanations

    Chinese mathematicians and painters

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

    The neuron primarily activates on numeric tokens—Arabic numerals and number‐containing strings (years, counts, digits, or math‐related number forms).

    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
     Lyman
    -0.71
    くださいね
    -0.69
    下单
    -0.68
     aerop
    -0.68
    하며
    -0.67
    Edgar
    -0.66
     Gaya
    -0.66
     seldom
    -0.66
     опла
    -0.66
    過去の
    -0.66
    POSITIVE LOGITS
    Ⓣ
    0.89
     Heron
    0.87
    DateUtils
    0.85
    Dio
    0.84
     Revolución
    0.82
     proposés
    0.81
     коммента
    0.80
    垛
    0.78
     memória
    0.78
     rhetorical
    0.77
    Activations Density 0.012%

    No Known Activations