© 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. 128579
    Prev
    Next
    INDEX
    Explanations

    This neuron detects occurrences of “establish” (and its inflected forms like “establishing” or “establishment”) and nearby phrases specifying what is being established.

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

    establishing

    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
     poderá
    -1.09
    そう
    -1.05
    前回
    -0.94
    ján
    -0.94
    Same
    -0.90
     подска
    -0.90
     そう
    -0.90
     sarà
    -0.89
     compliqué
    -0.89
    Referencies
    -0.88
    POSITIVE LOGITS
     establishes
    1.39
     establish
    1.37
     a
    1.19
    establish
    1.19
     for
    1.05
     establishing
    1.01
     unggul
    0.99
     firmly
    0.97
     pysty
    0.96
    ઃ
    0.95
    Activations Density 0.048%

    No Known Activations