© 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. 129478
    Prev
    Next
    INDEX
    Explanations

    This neuron detects Ruby on Rails controller code where a model object is instantiated and assigned to an instance variable (e.g. “@task = Task.new …”).

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

    Mam, Since, [ followed by specific nouns

    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
     hemd
    -1.13
     buhay
    -1.12
     női
    -1.12
     Vitamina
    -1.09
    зина
    -1.06
     muhimu
    -1.05
    melle
    -1.03
     iyon
    -1.00
    Substrate
    -1.00
    CEE
    -1.00
    POSITIVE LOGITS
     if
    1.30
     only
    1.27
     mostly
    1.13
     with
    1.07
     for
    1.05
     does
    1.05
    Fonto
    1.01
    Sú
    0.97
     mainly
    0.97
     or
    0.97
    Activations Density 0.007%

    No Known Activations