© 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. 39013
    Prev
    Next
    INDEX
    Explanations

    end of code statements

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

    The neuron is triggered by imperative code tokens—especially function or method calls (e.g. print, event‐complete callbacks) and return statements—in programming snippets.

    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
     then
    -1.20
     Then
    -0.93
    вік
    -0.91
    ключа
    -0.91
    wezen
    -0.90
     kemudian
    -0.88
    🪘
    -0.84
    bonsoir
    -0.84
    тных
    -0.83
     accedi
    -0.82
    POSITIVE LOGITS
     else
    1.00
    else
    0.91
    Evolution
    0.82
     ailleurs
    0.79
     voetbal
    0.79
    य़
    0.78
    URO
    0.77
    Jadi
    0.75
     Begründung
    0.75
     kebun
    0.75
    Activations Density 0.003%

    No Known Activations