© 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. 60155
    Prev
    Next
    INDEX
    Explanations

    first name, last name

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

    The neuron strongly activates on tokens that are named entities and numbers—e.g. personal names, organization names, years, scores, copyright notices—i.e. proper nouns and numeric identifiers.

    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
    ilson
    -0.82
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    -0.78
    Xls
    -0.77
     Socorro
    -0.77
    धारण
    -0.77
     uniquement
    -0.77
     ressemb
    -0.75
     turquoise
    -0.75
     <<<
    -0.74
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    -0.73
    POSITIVE LOGITS
     /\.
    0.74
    উ
    0.71
     к
    0.71
     級
    0.70
    yaret
    0.68
     получил
    0.68
    大赛
    0.68
     biệt
    0.67
     ugl
    0.67
    得到的
    0.66
    Activations Density 0.129%

    No Known Activations