© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38567
    Prev
    Next
    INDEX
    Explanations

    predicting words after expressions

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

    The neuron is a strong detector for numeric tokens (numerals and number literals, including integers and floats).

    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
    強調
    -0.96
     最後
    -0.93
     towering
    -0.91
     垃圾
    -0.88
     incompet
    -0.88
    踱
    -0.88
     作業
    -0.87
     dankbar
    -0.87
    Stage
    -0.85
    ování
    -0.85
    POSITIVE LOGITS
     llamó
    1.09
     duele
    1.09
     ponen
    1.09
    ệu
    1.08
    middels
    1.07
     sirva
    1.05
     tratan
    1.03
     sienta
    1.02
     wenden
    1.02
     siente
    1.02
    Activations Density 0.051%

    No Known Activations