© 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. 68421
    Prev
    Next
    INDEX
    Explanations

    initial-scale=1

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

    The neuron fires on numeric tokens (digit sequences), in particular the numbers specifying the viewport’s initial-scale in HTML meta tags.

    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
    verlet
    -1.23
    merken
    -1.23
    rechter
    -1.21
     srebr
    -1.20
    kennen
    -1.20
    wendige
    -1.19
    standen
    -1.16
     SECOND
    -1.16
     Sekret
    -1.16
     Onze
    -1.15
    POSITIVE LOGITS
     manifested
    1.40
     certainly
    1.35
     already
    1.34
     obviously
    1.30
     different
    1.25
     enormously
    1.25
     exclusively
    1.20
     roughly
    1.20
     perfectly
    1.18
     comprehensively
    1.16
    Activations Density 0.064%

    No Known Activations