© 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. 73870
    Prev
    Next
    INDEX
    Explanations

    variables with assignments

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

    The neuron broadly activates on numeric tokens, i.e. when the text contains numbers or numerical values.

    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
    ÄR
    -1.23
    多少钱
    -1.10
     people
    -1.05
    gså
    -1.02
     if
    -1.02
     مى
    -1.02
    cím
    -1.00
     will
    -0.99
    StatusOK
    -0.99
    }$.
    -0.98
    POSITIVE LOGITS
     zero
    1.09
    niño
    1.00
    providing
    0.97
    corazón
    0.96
     zmienia
    0.94
    =
    0.94
     zapewnia
    0.93
    ≳
    0.93
     สาม
    0.93
    similar
    0.91
    Activations Density 0.039%

    No Known Activations