© 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. 10-GEMMASCOPE-RES-131K
    7. 44688
    Prev
    Next
    INDEX
    Explanations

    per and units of measure

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

    The neuron fires on the word “per” (as used in rate/ratio expressions like “per month,” “per hour,” 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
            
    -2.02
    ↵↵
    -1.97
     but
    -1.95
     I
    -1.88
     same
    -1.84
     Voraussetzung
    -1.81
     in
    -1.77
     last
    -1.76
     fazendo
    -1.75
     following
    -1.73
    POSITIVE LOGITS
    providedIn
    2.44
    Setelah
    2.19
    Setiap
    2.14
    оргаш
    2.09
    蹓
    2.08
    Ketika
    1.98
     dezelve
    1.98
    Selain
    1.96
    卨
    1.95
     valentín
    1.95
    Activations Density 0.010%

    No Known Activations