© 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. 10-GEMMASCOPE-RES-131K
    7. 65368
    Prev
    Next
    INDEX
    Explanations

    comparisons and programming tests

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

    The neuron strongly activates on the Chinese characters “计” and “较” (as in the word 计较, “to calculate/compare”), i.e. tokens marking a calculation/comparison term in Chinese.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     of
    -2.48
     validés
    -2.13
     sowohl
    -2.13
     mestre
    -2.11
    ísima
    -2.09
    ৪
    -2.09
    🤍
    -2.08
     believes
    -2.05
     dezelfde
    -2.05
    ?"
    -2.03
    POSITIVE LOGITS
    '
    3.20
    2
    2.73
     erhi
    2.73
    4
    2.70
    7
    2.55
    这样的
    2.45
    1
    2.44
    9
    2.41
     evitando
    2.38
    5
    2.22
    Activations Density 0.001%

    No Known Activations