© 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. 65738
    Prev
    Next
    INDEX
    Explanations

    foreign language or loanwords

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

    The neuron flags content‐bearing (mostly longer or less common) words in the running prose, as opposed to headings, punctuation, or very common stop–words.

    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
    ();
    -0.92
     promise
    -0.88
     Dane
    -0.82
     deter
    -0.82
    promise
    -0.77
    んこ
    -0.76
    зите
    -0.75
     Bunk
    -0.73
    さで
    -0.72
     triplicate
    -0.70
    POSITIVE LOGITS
     avantaj
    1.02
    thered
    0.94
    ᱶ
    0.86
     asalto
    0.84
     accustomed
    0.83
    appartamento
    0.82
    THANK
    0.82
     bandit
    0.82
     着物
    0.81
    formatie
    0.81
    Activations Density 0.029%

    No Known Activations