© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 131065
    Prev
    Next
    INDEX
    Explanations

    The neuron is looking for numeric tokens—particularly multi‐digit numbers, decimals, and other quantitative values embedded in the text.

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

    statistics following noun

    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
    簡介
    -0.88
     pouze
    -0.85
     erklär
    -0.84
     説明
    -0.81
     usually
    -0.77
    こんばんは
    -0.77
    ddf
    -0.76
    apparence
    -0.76
     fehlen
    -0.76
     vorigen
    -0.75
    POSITIVE LOGITS
     below
    1.34
    Below
    1.30
    <table>
    1.21
    below
    1.16
     Below
    1.11
     voici
    1.01
     berikut
    1.01
    は以下の
    0.99
    Berikut
    0.97
     Voici
    0.96
    Activations Density 0.090%

    No Known Activations