© 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. 83153
    Prev
    Next
    INDEX
    Explanations

    the US, USA, UK

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

    The neuron consistently activates on tokens that are pure numbers (digit sequences), i.e. it fires whenever it sees numeric data like counts, years, or other statistics.

    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
     thing
    -1.05
     other
    -0.96
    ของ
    -0.90
    くださいませ
    -0.84
    包裹
    -0.83
    蓝色
    -0.82
    我是
    -0.82
     incubated
    -0.81
    هرة
    -0.81
    撷
    -0.81
    POSITIVE LOGITS
     US
    2.28
     UK
    1.39
     U
    1.37
     EU
    1.03
     USA
    1.02
     अमे
    1.01
    🚶
    0.96
    US
    0.95
     américains
    0.94
    addressed
    0.93
    Activations Density 0.063%

    No Known Activations