© 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. 22-GEMMASCOPE-RES-131K
    4. 12928
    Prev
    Next
    INDEX
    Explanations

    trigonometric functions and variables

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

    The neuron activates on in-text mathematical expressions—especially trig functions (like sin, cos, sec, tan) and Greek-letter variables (like θ, φ, Ψ).

    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
     iniziato
    -1.35
     после
    -1.27
     dichiarato
    -1.23
     protects
    -1.20
    又は
    -1.18
    👡
    -1.17
     spese
    -1.17
     nemico
    -1.14
     lunghe
    -1.13
    produzione
    -1.13
    POSITIVE LOGITS
    ↵↵
    1.20
    違って
    1.07
    おもちゃ
    1.05
     trá
    1.00
     английском
    0.98
    putting
    0.97
    
    0.96
     étoiles
    0.94
     attente
    0.94
     riv
    0.94
    Activations Density 0.004%

    No Known Activations