© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 130814
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively activates on occurrences of the code‐token “String,” i.e. it picks out String type declarations.

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

    code variable declarations and assignments

    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
    鬟
    -1.23
     =>
    
    -1.18
     Pueden
    -1.16
    zzet
    -1.14
    fordshire
    -1.08
     Ofrece
    -1.08
    дай
    -1.06
     Até
    -1.06
     Lleva
    -1.06
    tonsoft
    -1.05
    POSITIVE LOGITS
    のよ
    1.07
     may
    1.05
    しましたが
    1.03
     it
    1.01
    kranz
    0.99
     could
    0.98
    的吧
    0.96
     aspects
    0.96
     value
    0.94
    置いて
    0.92
    Activations Density 0.000%

    No Known Activations