© 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. 10855
    Prev
    Next
    INDEX
    Explanations

    there's the [noun phrase]

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

    The neuron strongly activates on instances of the definite article “the” (including sentence‐initial “The”), effectively detecting that common article token.

    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
     this
    -2.17
     sommige
    -1.72
     it
    -1.70
     its
    -1.66
    ↵↵
    -1.62
    ophageal
    -1.58
     verdieping
    -1.57
     which
    -1.55
     lucre
    -1.53
     使っ
    -1.51
    POSITIVE LOGITS
     occasional
    1.74
    𐍆
    1.70
     efekt
    1.64
     και
    1.60
     ducha
    1.59
     dát
    1.59
     „
    1.59
     ซึ่ง
    1.55
     unforgettable
    1.53
     Which
    1.52
    Activations Density 0.013%

    No Known Activations