© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10890
    Prev
    Next
    INDEX
    Explanations

    names and surnames

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

    The neuron fires on tokens whose letters alternate vowel-consonant (or consonant-vowel) over several positions—e.g. Ariana, Mariah, Prairie, Aurora, Cyrus.

    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
    la
    -2.45
    ”.
    -2.27
     However
    -2.25
    也
    -2.17
     This
    -2.17
     Nevertheless
    -2.05
    ........
    -2.03
    li
    -2.03
     einzigen
    -2.00
    It
    -1.97
    POSITIVE LOGITS
    壜
    2.33
     coté
    2.14
    わたしは
    2.06
    鹀
    2.03
     luminoso
    1.95
     abiye
    1.95
    过去的
    1.93
    粜
    1.88
    昔の
    1.86
     torba
    1.84
    Activations Density 0.003%

    No Known Activations