© 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. 22-GEMMASCOPE-RES-131K
    7. 67364
    Prev
    Next
    INDEX
    Explanations

    square roots and mathematical operations

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

    The neuron is detecting numeric literals (standalone numbers or digit sequences) within the 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
     herbes
    -1.06
     maal
    -0.99
    Σε
    -0.94
    тров
    -0.93
    コウ
    -0.92
     %)
    -0.89
    γραφία
    -0.88
     provare
    -0.88
     strate
    -0.87
     papas
    -0.85
    POSITIVE LOGITS
     imple
    0.90
     comprometido
    0.89
     suela
    0.88
     fame
    0.86
    graceful
    0.85
    rewsbury
    0.85
     Zwar
    0.85
    forbidden
    0.84
    вік
    0.84
    Fame
    0.83
    Activations Density 0.006%

    No Known Activations