© 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. 64847
    Prev
    Next
    INDEX
    Explanations

    first or beginning

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

    The neuron fires on isolated code‐style tokens—things like identifiers, filenames or parameters containing digits, underscores or mixed‐case—rather than ordinary English words.

    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
     sterke
    -0.95
    ためには
    -0.95
    🐬
    -0.91
    課程
    -0.90
    
    -0.90
    ęb
    -0.90
     carreg
    -0.87
     TOPICS
    -0.85
     problemet
    -0.85
    իր
    -0.85
    POSITIVE LOGITS
     first
    1.19
     with
    1.00
     for
    0.86
     Erster
    0.82
     yes
    0.81
    ríamos
    0.80
     Mungkin
    0.77
     on
    0.76
     пър
    0.75
     beginning
    0.75
    Activations Density 0.046%

    No Known Activations