© 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. 73061
    Prev
    Next
    INDEX
    Explanations

    process and apparatus

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

    The neuron fires strongly whenever the text refers to a “process” (especially in patent‐style descriptions of methods).

    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
    鱘
    -1.73
    刧
    -1.68
     plát
    -1.66
     tubos
    -1.65
     électron
    -1.63
    ⢧
    -1.61
    ⢅
    -1.61
    ⃤
    -1.60
     vlas
    -1.59
     symboles
    -1.59
    POSITIVE LOGITS
     masih
    1.59
    .
    1.45
     Meksiko
    1.38
    ↵
    1.38
     Still
    1.37
     is
    1.35
    ↵↵
    1.31
     sebagian
    1.30
     meinen
    1.30
     Making
    1.30
    Activations Density 0.088%

    No Known Activations