© 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. 12993
    Prev
    Next
    INDEX
    Explanations

    URL-encoded text and identifiers

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

    The neuron activates on numeric literals (whole numbers and decimals).

    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
     Euró
    -0.92
    营销
    -0.88
     kó
    -0.87
     suéter
    -0.85
     kré
    -0.84
    ↠
    -0.84
    mohon
    -0.84
     hermé
    -0.82
     euró
    -0.81
     projek
    -0.81
    POSITIVE LOGITS
     said
    0.96
     anyway
    0.88
    úncia
    0.86
    守护
    0.84
    rollment
    0.84
     informaron
    0.83
    ihnachts
    0.82
     peuvent
    0.82
     partenza
    0.82
    都可以
    0.81
    Activations Density 0.032%

    No Known Activations