© 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. 71731
    Prev
    Next
    INDEX
    Explanations

    code structures

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

    The neuron predominantly fires on numeric tokens—e.g. literal numbers, array‐sizes, numeric constants or digits in the code.

    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
    curs
    -0.85
     kelapa
    -0.85
    раль
    -0.82
     koning
    -0.82
    ᵁ
    -0.82
     castillo
    -0.81
    ஒ
    -0.81
     mentales
    -0.81
     муніципалі
    -0.80
    iene
    -0.80
    POSITIVE LOGITS
    jaardag
    0.93
    𝑀
    0.86
     ziyaretçi
    0.85
     giàu
    0.84
     سيارة
    0.79
     =\
    0.78
    鋲
    0.77
     mówi
    0.77
    理石
    0.76
    使った
    0.75
    Activations Density 0.028%

    No Known Activations