© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26280
    Prev
    Next
    INDEX
    Explanations

    code and titles

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

    The neuron fires on tokens ending with the letters “se.”

    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
     their
    -1.31
    
    -1.28
     Japão
    -1.24
     Acesso
    -1.13
    子さん
    -1.11
     refusé
    -1.09
     conftest
    -1.04
     berdua
    -1.04
     mitte
    -1.04
     different
    -1.04
    POSITIVE LOGITS
     exemplu
    1.16
     erwor
    1.12
     horrid
    1.12
     culoare
    1.10
     vorhandenen
    1.10
     zugel
    1.09
     ardent
    1.08
    を作り
    1.08
     loca
    1.04
     vorhand
    1.03
    Activations Density 0.000%

    No Known Activations