© 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. 115679
    Prev
    Next
    INDEX
    Explanations

    words starting with L

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

    The neuron detects romanized Indic-language words or morphemes (i.e. transliterations of Hindi/Urdu terms and names).

    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
     leksikon
    -0.95
     syk
    -0.88
     museo
    -0.85
    轼
    -0.84
     prata
    -0.82
    pref
    -0.81
    どり
    -0.76
    がありますが
    -0.76
     meren
    -0.76
     tien
    -0.74
    POSITIVE LOGITS
     ल
    0.90
     lány
    0.90
    に似
    0.84
     SEG
    0.83
    ismiss
    0.81
    Estimator
    0.77
     tych
    0.76
     waving
    0.76
    Thousands
    0.75
    Target
    0.74
    Activations Density 0.006%

    No Known Activations