© 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. 70594
    Prev
    Next
    INDEX
    Explanations

    daily mail publication

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

    The neuron is looking for numeric expressions and related tokens—such as numbers, dates, measurements, hyphenated counts or ranges, and percentages.

    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
    ISMO
    -1.15
     vuitton
    -1.02
    BYTE
    -0.98
     metus
    -0.98
     diesen
    -0.98
    🤟
    -0.98
    킹
    -0.96
    example
    -0.95
    💓
    -0.93
     of
    -0.92
    POSITIVE LOGITS
    dailymail
    1.43
    ⠘
    1.06
    ￡
    1.06
     vergangenen
    1.05
    нией
    1.05
    inspirational
    1.05
    adorable
    1.03
     سابق
    1.02
    已經
    1.02
     poucos
    1.02
    Activations Density 0.006%

    No Known Activations