© 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. 22-GEMMASCOPE-RES-131K
    7. 66159
    Prev
    Next
    INDEX
    Explanations

    time intervals and frequencies

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

    The neuron detects numeric tokens or quantities (e.g. counts, amounts, time intervals, or other numbers) in text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     Schauspielerin
    -1.03
    加坡
    -0.91
    为止
    -0.90
    ḻ
    -0.87
    didReceive
    -0.87
     играет
    -0.87
     article
    -0.85
     options
    -0.83
    なども
    -0.82
    憊
    -0.81
    POSITIVE LOGITS
    getParent
    1.10
    每
    1.08
     schuldig
    1.04
     during
    1.00
     infrequent
    1.00
    ทุก
    0.98
    urende
    0.97
     milestone
    0.96
     düğ
    0.95
     Frequency
    0.94
    Activations Density 0.003%

    No Known Activations