© 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. 116020
    Prev
    Next
    INDEX
    Explanations

    rounding to the nearest

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

    This neuron fires on phrases that instruct to round numbers—e.g. “rounded to … decimal places” or “rounded to the nearest ….”

    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
     bardzo
    -0.82
    冪
    -0.81
     практически
    -0.77
     rebajas
    -0.75
     especially
    -0.73
     velmi
    -0.71
    Designation
    -0.70
     очень
    -0.69
     размеры
    -0.67
    軽減
    -0.67
    POSITIVE LOGITS
     nearest
    2.44
    nearest
    1.99
    Nearest
    1.93
     Nearest
    1.67
     closest
    1.48
     najbli
    1.43
     ближай
    1.31
    Closest
    1.30
    closest
    1.22
     tenth
    1.12
    Activations Density 0.019%

    No Known Activations