© 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. 34-GEMMASCOPE-RES-131K
    7. 31964
    Prev
    Next
    INDEX
    Explanations

    code is regenerated

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

    The neuron looks for autogenerated‐code warning or “do not edit” comments—specifically the boilerplate about lost changes if the code is regenerated.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     gyóg
    -0.92
    ǰ
    -0.90
     reaksi
    -0.88
     香水
    -0.88
     Vila
    -0.88
    egl
    -0.88
    telah
    -0.83
    żer
    -0.83
    سیم
    -0.81
    اولة
    -0.81
    POSITIVE LOGITS
     regenerate
    1.09
     regenerated
    1.05
     generated
    1.04
     regener
    0.94
     Regener
    0.90
     다시
    0.90
    regener
    0.85
     behavi
    0.83
    valuate
    0.81
     volverá
    0.80
    Activations Density 0.012%

    No Known Activations