© 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. 10-GEMMASCOPE-RES-131K
    7. 19617
    Prev
    Next
    INDEX
    Explanations

    Persistent symptomatic

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

    The neuron consistently activates on isolated “X” (or “x”) characters when they appear as part of abbreviations or acronyms (e.g. LOX, LOXs, ITx).

    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
     toont
    -1.68
    idade
    -1.64
     interieur
    -1.58
     bague
    -1.57
     lois
    -1.55
     счастье
    -1.55
     chaussure
    -1.54
    wheres
    -1.53
     bouteille
    -1.52
     passagers
    -1.52
    POSITIVE LOGITS
     Wirken
    1.81
    1
    1.76
    Our
    1.70
     też
    1.69
    也会
    1.64
     Umfeld
    1.60
    wiście
    1.55
     också
    1.52
     Sekarang
    1.52
     Therefore
    1.52
    Activations Density 0.000%

    No Known Activations