© 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. 59850
    Prev
    Next
    INDEX
    Explanations

    per unit of measure

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

    The neuron strongly activates on the word “per” when used to express rates or unit‐based quantities (e.g. “per user,” “per second,” “per mailbox,” etc.).

    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
    ↵↵
    -2.36
    “
    -2.31
     \&
    -2.19
    ౪
    -2.08
    »
    -1.92
     THE
    -1.88
    [
    -1.87
    但他
    -1.86
    但她
    -1.84
    {
    -1.84
    POSITIVE LOGITS
    瞜
    2.27
    Добавлено
    2.16
    erintah
    2.16
    鶲
    2.09
    Скачать
    2.06
     them
    2.02
     土産
    2.00
    Привет
    1.95
    llavero
    1.92
    んですが
    1.88
    Activations Density 0.011%

    No Known Activations