© 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. 34-GEMMASCOPE-RES-131K
    7. 115792
    Prev
    Next
    INDEX
    Explanations

    XML resource strings and multilingual labels

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

    The neuron fires on numeric tokens (digits or number sequences) such as XML attribute values, resource IDs, permission codes, etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     styles
    -0.87
    嗔
    -0.77
    ಸ್
    -0.75
     regras
    -0.75
     recetas
    -0.75
    tema
    -0.74
     estu
    -0.71
    苫
    -0.71
     Styles
    -0.70
     luk
    -0.70
    POSITIVE LOGITS
    大致
    0.98
     anglès
    0.95
     …………
    0.92
     multilingual
    0.86
     translators
    0.84
    Mirrors
    0.82
     úpl
    0.82
     string
    0.80
     parro
    0.79
     worded
    0.79
    Activations Density 0.001%

    No Known Activations