© 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. 34-GEMMASCOPE-RES-131K
    7. 65289
    Prev
    Next
    INDEX
    Explanations

    words starting with prag, trag, frag, brag

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

    The neuron fires on any appearance of the letter sequence “rag” (as in “frag,” “trag,” “prag,” etc.) embedded inside a word.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     garmin
    -0.83
     
    -0.81
     בישראל
    -0.78
    nání
    -0.78
    RUB
    -0.77
     insbesondere
    -0.77
    MONG
    -0.77
     gour
    -0.77
     epson
    -0.76
     itunes
    -0.75
    POSITIVE LOGITS
    corresponding
    0.85
    aluz
    0.78
     Alors
    0.77
    原
    0.77
     corresponding
    0.75
    ⚌
    0.75
    wanted
    0.73
     King
    0.73
     odeur
    0.72
    PROG
    0.71
    Activations Density 0.014%

    No Known Activations