© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 66656
    Prev
    Next
    INDEX
    Explanations

    numbers and punctuation followed by words

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

    The neuron fires on standalone numeric tokens (dates, street numbers, counts) and news‐style dateline/location tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     http
    -1.54
     xrange
    -1.52
    componentWill
    -1.35
    http
    -1.34
    されていた
    -1.20
     especially
    -1.15
    不下
    -1.13
    ``,
    -1.13
    ISIS
    -1.13
     their
    -1.13
    POSITIVE LOGITS
    androidx
    1.87
     TikTok
    1.86
     NFTs
    1.74
     covid
    1.70
     COVID
    1.67
    . 
    1.63
     Biden
    1.61
     ۱۴۰۰
    1.60
     ChatGPT
    1.60
     Covid
    1.60
    Activations Density 0.011%

    No Known Activations