© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 71196
    Prev
    Next
    INDEX
    Explanations

    leave absence furlough loan

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

    The neuron detects formal references to temporary absences or transfers—especially words like “leave,” “absence,” and “loan.”

    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
     spéciales
    -1.28
    полне
    -1.26
     tertawa
    -1.23
     spolup
    -1.19
     complètes
    -1.17
     perros
    -1.17
     bemerken
    -1.16
     atrav
    -1.15
     velikost
    -1.14
     verhind
    -1.13
    POSITIVE LOGITS
     from
    1.52
    ворю
    1.32
    '...
    1.23
     FOR
    1.16
     наш
    1.16
    측
    1.16
     "
    1.16
     out
    1.15
     such
    1.15
     l
    1.14
    Activations Density 0.030%

    No Known Activations