© 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. 65846
    Prev
    Next
    INDEX
    Explanations

    country codes

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

    The neuron flags isolated identifier‐style tokens—short all‐caps strings and numbers used as codes, IDs, reporting marks, or other discrete alphanumeric identifiers.

    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
     What
    -1.31
     even
    -1.25
    what
    -1.23
    !!!!
    -1.13
     Even
    -1.13
    ........
    -1.13
     The
    -1.09
    .......
    -1.09
     How
    -1.07
    !!!!!
    -1.07
    POSITIVE LOGITS
     やすい
    1.45
    Avec
    1.39
     refuted
    1.38
    pendentif
    1.38
    Tout
    1.36
     cóc
    1.35
    
    1.34
    
    1.33
     ESTADO
    1.32
    
    1.30
    Activations Density 0.002%

    No Known Activations