© 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. 70921
    Prev
    Next
    INDEX
    Explanations

    shipwrecked or stranded

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

    The neuron fires on past-participles of misfortune verbs—especially nautical ones—such as stranded, wrecked, and beached.

    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
    o
    -2.02
    6
    -1.68
     .
    -1.52
    4
    -1.48
     appelé
    -1.48
     fevereiro
    -1.46
    ”،
    -1.46
    re
    -1.45
    ков
    -1.42
    ⑵
    -1.42
    POSITIVE LOGITS
    瘓
    1.73
    霙
    1.60
     on
    1.53
     stranded
    1.52
     übri
    1.48
    采用了
    1.45
    我們
    1.45
     Putih
    1.43
    谘
    1.43
    putan
    1.42
    Activations Density 0.014%

    No Known Activations