© 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. 22-GEMMASCOPE-RES-131K
    7. 130018
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly activates on the word “The” when it begins a sentence that presents numeric measurements or other quantified data.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    The followed by descriptor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     célè
    -1.77
     pérd
    -1.55
     rése
    -1.54
     éta
    -1.54
    甗
    -1.48
    ishment
    -1.48
    쐈
    -1.48
    €”
    -1.47
     fré
    -1.46
    鶿
    -1.46
    POSITIVE LOGITS
     also
    2.34
     other
    1.84
     others
    1.83
     on
    1.70
     one
    1.55
     more
    1.48
     two
    1.47
     most
    1.44
     only
    1.41
     three
    1.38
    Activations Density 0.040%

    No Known Activations