© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 14205
    Prev
    Next
    INDEX
    Explanations

    Ruby code structures

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

    The neuron fires specifically on the pipe character “|” when it’s used to delimit block parameters in Ruby code.

    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
     and
    -1.96
    
    -1.61
    猡
    -1.56
    TRIBUN
    -1.55
    Puedo
    -1.54
    <bos>
    -1.53
     When
    -1.46
     широкий
    -1.45
    資料來源
    -1.43
     to
    -1.40
    POSITIVE LOGITS
     mètre
    2.05
    熥
    1.77
    ఽ
    1.75
     logros
    1.69
     beobachtet
    1.68
     lessen
    1.67
     lesquelles
    1.66
     חיצוני
    1.58
    ,“
    1.57
     ukry
    1.56
    Activations Density 0.001%

    No Known Activations