© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14094
    Prev
    Next
    INDEX
    Explanations

    phrase, concept, or fad

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

    The neuron specializes in spotting discussions or mentions of “phrases” (especially catchphrases or recurring lines).

    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
     wybrać
    -0.88
     Фер
    -0.87
    äfte
    -0.84
    omy
    -0.84
    ardino
    -0.82
     възможно
    -0.82
     dokładnie
    -0.81
     варіан
    -0.80
     wynika
    -0.80
    oxane
    -0.79
    POSITIVE LOGITS
     phenomena
    1.14
     distinctive
    1.14
     signature
    1.09
     ubiquitous
    1.02
     noved
    1.02
     viral
    0.96
     повторя
    0.96
     invented
    0.96
     фир
    0.95
    foundland
    0.95
    Activations Density 0.124%

    No Known Activations