© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 30613
    Prev
    Next
    INDEX
    Explanations

    numbers and time units

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

    The neuron is tuned to pick out numeric tokens and numeric literals (e.g. standalone numbers, digits, timestamps, or numeric values in code/config).

    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
     minute
    -1.09
    minute
    -1.00
     minutes
    -0.91
     milliseconds
    -0.84
     ADVIS
    -0.81
    女王
    -0.79
    分鐘
    -0.76
     минуту
    -0.76
    Minute
    -0.76
     seconde
    -0.73
    POSITIVE LOGITS
     ten
    0.99
     five
    0.98
     zehn
    0.91
     fifteen
    0.79
     fünf
    0.78
    five
    0.77
    vertisement
    0.77
     eight
    0.75
    excl
    0.73
     Maud
    0.73
    Activations Density 0.038%

    No Known Activations