© 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. 34161
    Prev
    Next
    INDEX
    Explanations

    code syntax or political figures

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

    The neuron fires on code tokens that signal control flow or logical operations (e.g. “then,” “&&,” “||,” template‐conditional markers, etc.).

    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
    درب
    -1.05
     fantasma
    -1.02
     kny
    -1.01
     semoga
    -1.00
    ável
    -0.97
    idão
    -0.96
     Politique
    -0.94
     umur
    -0.94
     teatral
    -0.93
     lehet
    -0.92
    POSITIVE LOGITS
    /***
    0.93
     AOS
    0.88
     процесса
    0.88
     Helico
    0.87
    IJ
    0.84
     telefonisch
    0.84
    u
    0.84
     homogenous
    0.84
    AppConfig
    0.83
     MIST
    0.83
    Activations Density 0.001%

    No Known Activations