© 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. 34-GEMMASCOPE-RES-131K
    7. 1865
    Prev
    Next
    INDEX
    Explanations

    soil and cultivation

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

    references to soil, earth, or ground in agricultural or environmental contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    Explanation of neuron 4 behavior: the main thing this neuron does is find common function words—especially high-frequency prepositions and articles (e.g. of, in, to, the).

    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
    粵
    -0.76
    indow
    -0.73
     nạ
    -0.73
    irce
    -0.69
    💦
    -0.68
    PROOF
    -0.68
     Rapper
    -0.67
     LCA
    -0.67
    listView
    -0.66
    xj
    -0.66
    POSITIVE LOGITS
     soil
    5.41
     soils
    3.69
    soil
    3.66
     Soil
    3.56
    Soil
    3.50
     SOIL
    2.94
    土壤
    2.69
     Soils
    2.59
    土
    2.56
    SOIL
    2.47
    Activations Density 0.042%

    No Known Activations