© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128373
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on technical, system‐oriented terms—words naming components or states of a computing environment (e.g. “my,” “test,” “machine,” “running,” “dynamic,” “nodes”).

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

    dynamic hierarchies or layers

    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
    だいたい
    -0.88
     zowel
    -0.85
    んだろう
    -0.85
     일반
    -0.84
     ListView
    -0.84
     aproximativ
    -0.84
     ayrıca
    -0.83
     Splitter
    -0.82
     pleaf
    -0.80
     Elem
    -0.79
    POSITIVE LOGITS
     but
    1.37
     if
    1.33
     with
    1.17
     without
    1.05
     Etik
    1.03
     nhưng
    0.93
    But
    0.92
     too
    0.88
     pitcher
    0.87
     when
    0.85
    Activations Density 0.137%

    No Known Activations