© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65215
    Prev
    Next
    INDEX
    Explanations

    formatted text or data structures

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

    This neuron responds to occurrences of the word “formatted.”

    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
    ﮏ
    -2.77
     lidí
    -2.70
     नए
    -2.59
     potrebné
    -2.59
     vlastnosti
    -2.53
     nových
    -2.50
     člán
    -2.41
    嵛
    -2.39
     lilies
    -2.39
    嚒
    -2.36
    POSITIVE LOGITS
    けれ
    2.73
    ーーーー
    2.61
    
    2.59
    別人
    2.58
    …”
    2.53
    To
    2.48
    但她
    2.45
    as
    2.44
    For
    2.42
    とはいえ
    2.38
    Activations Density 0.002%

    No Known Activations