© 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. 34-GEMMASCOPE-RES-131K
    7. 34513
    Prev
    Next
    INDEX
    Explanations

    Annie and names

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

    The neuron is picking out named-entity–style tokens—especially proper names (like “Annie …”) and related numeric/date 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_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
     death
    -0.72
    母亲
    -0.71
    versible
    -0.70
     appena
    -0.70
     collapse
    -0.70
    SMR
    -0.69
    兄
    -0.69
    嚴
    -0.69
    sibling
    -0.68
    ulata
    -0.67
    POSITIVE LOGITS
     Annie
    0.97
    Annie
    0.86
     پرورش
    0.73
    姝
    0.72
    マイナス
    0.71
     інтер
    0.71
     đấu
    0.70
     ANN
    0.70
    Cells
    0.70
     BERN
    0.69
    Activations Density 0.010%

    No Known Activations