© 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. 10-GEMMASCOPE-RES-131K
    7. 10902
    Prev
    Next
    INDEX
    Explanations

    bending beams and radius

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

    The neuron strongly activates on technical structural‐engineering terms and measurements—especially words describing bending or flexural tests and load-bearing components (e.g. bending, flexural, beam, spar).

    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
    </u>
    -1.45
    </strong>
    -1.40
     a
    -1.38
    讪
    -1.31
    gelegen
    -1.27
     (
    -1.25
     factura
    -1.25
    wendige
    -1.22
     dekoratif
    -1.22
    ズニー
    -1.22
    POSITIVE LOGITS
    étranger
    1.48
     our
    1.39
    荞
    1.34
     PRESENTED
    1.33
     compartilhar
    1.30
    涐
    1.26
    笵
    1.23
    っけ
    1.20
     PARTNERS
    1.20
     سامسون
    1.20
    Activations Density 0.082%

    No Known Activations