© 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. 102859
    Prev
    Next
    INDEX
    Explanations

    numbers and digits

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

    The neuron activates on the large numeric tokens used in unified-diff hunk headers (the line-number and range numbers in patch metadata).

    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
     textil
    -1.11
     metais
    -1.09
     blusa
    -1.05
    ----|
    -1.05
    ---|
    -1.03
     fré
    -1.03
    bewerken
    -1.00
     atos
    -1.00
     Temperatur
    -0.98
     rosas
    -0.98
    POSITIVE LOGITS
    3
    1.12
    k
    0.97
    活用
    0.95
    6
    0.93
    8
    0.93
    そういった
    0.92
     meravigli
    0.90
    9
    0.90
     numai
    0.88
    7
    0.88
    Activations Density 0.003%

    No Known Activations