© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26879
    Prev
    Next
    INDEX
    Explanations

    code assignments and sql

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

    This neuron activates on code identifier names (variables or parameters) on the left‐hand side of assignments or declarations.

    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
    leicht
    -1.07
    kenal
    -0.98
     mariée
    -0.98
    ltä
    -0.95
    класс
    -0.95
    ühm
    -0.92
     jotka
    -0.92
     กล่อง
    -0.90
    madre
    -0.90
     Namun
    -0.89
    POSITIVE LOGITS
     next
    1.10
     навіть
    1.06
    urator
    0.98
    plotlib
    0.97
     even
    0.96
     than
    0.94
     said
    0.93
    さらに
    0.93
    Lmfao
    0.91
    داً
    0.91
    Activations Density 0.001%

    No Known Activations