© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 118112
    Prev
    Next
    INDEX
    Explanations

    technical component identifiers

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

    The neuron fires on uppercase code tokens and numeric literals—e.g. macro/constant names, all-caps identifiers, and digits in programming definitions.

    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
     garb
    -0.87
     BOLD
    -0.81
    Lá
    -0.78
     examining
    -0.78
    Scary
    -0.77
    考察
    -0.77
    教學
    -0.76
    ciler
    -0.75
    alang
    -0.75
    Lí
    -0.75
    POSITIVE LOGITS
     PHYSICS
    0.93
     Bitmap
    0.90
    癇
    0.89
    binaan
    0.85
     Quaternion
    0.84
     sumber
    0.84
     SwiftUI
    0.82
     Lennon
    0.81
    Coding
    0.81
    ложение
    0.80
    Activations Density 0.008%

    No Known Activations