© 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. 73119
    Prev
    Next
    INDEX
    Explanations

    countries and regions

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

    The neuron selectively activates on capitalized tokens—including country codes, proper nouns, and acronyms.

    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
    owohl
    -1.12
     those
    -0.95
    mischer
    -0.94
    خواهد
    -0.94
    其中
    -0.93
    iecie
    -0.93
     adonan
    -0.91
    uskas
    -0.89
     affatto
    -0.88
    蜜
    -0.86
    POSITIVE LOGITS
    amazing
    1.01
     hacerse
    1.00
    譁
    0.98
    everything
    0.98
     wartet
    0.95
    Dette
    0.93
    矚
    0.93
     basiert
    0.92
    Hvis
    0.91
     senaste
    0.91
    Activations Density 0.163%

    No Known Activations