© 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. 128157
    Prev
    Next
    INDEX
    Explanations

    The neuron is triggered by programming‐language header tokens (e.g. the “:” in file‐label lines and “import” statements) in code snippets.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    African presidents

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     specifik
    -1.08
     koncentr
    -1.01
     MARCH
    -0.98
     balkon
    -0.96
     blusa
    -0.93
    pleft
    -0.93
     GUYS
    -0.92
     mies
    -0.92
     produkt
    -0.91
     estante
    -0.91
    POSITIVE LOGITS
     coronav
    0.98
     ながら
    0.93
     MHC
    0.92
     embers
    0.88
     navigable
    0.88
    乗車
    0.86
    getL
    0.85
     wass
    0.84
     renegade
    0.84
     entw
    0.84
    Activations Density 0.000%

    No Known Activations