© 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. 99807
    Prev
    Next
    INDEX
    Explanations

    bending and lowering movements

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

    This neuron fires on participial verbs describing low‐to‐the‐ground or bent body postures or movements (e.g., crouching, squatting, kneeling, stooping, bending, leaning).

    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
     дорогой
    -1.97
     helft
    -1.95
    轳
    -1.94
     cheminée
    -1.77
    я
    -1.77
    跄
    -1.70
    ⟢
    -1.69
    战争
    -1.66
     都
    -1.63
     Hacer
    -1.62
    POSITIVE LOGITS
    ’
    1.97
    .”
    1.79
     ).
    1.74
     '
    1.73
     jLabel
    1.73
     structures
    1.66
    ,”
    1.66
    ”,
    1.64
    外表
    1.63
     beneath
    1.61
    Activations Density 0.004%

    No Known Activations