© 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. 10-GEMMASCOPE-RES-131K
    7. 38870
    Prev
    Next
    INDEX
    Explanations

    Rutledge, Rutgers, doctype html

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

    The neuron fires on the character sequence “Rut” (e.g. in Rutledge, Rutana, Rutgers), i.e. tokens containing “Rut.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    铼
    -3.14
     онъ
    -3.08
     betekent
    -2.94
    
    -2.94
    鲣
    -2.88
     zlep
    -2.84
    鸲
    -2.84
     dekoratif
    -2.81
    ﻒ
    -2.80
     voorbeeld
    -2.75
    POSITIVE LOGITS
    1
    3.77
    2
    3.73
    3
    3.42
    x
    3.36
      
    3.30
    p
    3.23
    6
    3.05
    8
    3.02
    <i>
    3.02
    5
    2.91
    Activations Density 0.002%

    No Known Activations