© 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. 34-GEMMASCOPE-RES-131K
    4. 97887
    Prev
    Next
    INDEX
    Explanations

    Daily News

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

    The neuron activates on tokens that contain digits (e.g. version numbers, years, or numeric literals).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     coarser
    -0.88
     клавиа
    -0.79
     entities
    -0.78
     nodal
    -0.78
     global
    -0.76
     niestety
    -0.75
     uninten
    -0.74
     globally
    -0.74
     ELEMENTS
    -0.74
    Merchant
    -0.73
    POSITIVE LOGITS
    ˂
    0.88
     stump
    0.79
    են
    0.76
    蚕
    0.75
     ebbe
    0.75
    apse
    0.75
     cren
    0.75
    眙
    0.75
     سرمایه
    0.74
    ERSON
    0.73
    Activations Density 0.002%

    No Known Activations