© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 127985
    Prev
    Next
    INDEX
    Explanations

    Dutch and German phrases and names

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

    The neuron fires strongly on numeric tokens—i.e. digits, dates, times, and other number values in running 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_10/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
    嘿嘿
    -1.50
    ".
    -1.42
    锨
    -1.41
    罍
    -1.39
    姈
    -1.38
     suport
    -1.38
    particularly
    -1.38
     verra
    -1.36
     però
    -1.35
     они
    -1.34
    POSITIVE LOGITS
    schedulers
    1.48
            
    1.45
    zegovina
    1.43
     παρου
    1.41
     one
    1.36
     system
    1.35
    比率
    1.35
     prevalent
    1.35
     people
    1.34
     article
    1.34
    Activations Density 0.001%

    No Known Activations