© 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. 38278
    Prev
    Next
    INDEX
    Explanations

    proper names and titles

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

    The neuron detects capitalized proper nouns and named entities (e.g. people’s names, place names, or titles).

    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
    WHIT
    -0.83
    ═
    -0.73
    headers
    -0.73
    泛
    -0.73
     pinch
    -0.73
     tune
    -0.72
     tunes
    -0.71
     Livingstone
    -0.71
     CIR
    -0.71
    クトル
    -0.70
    POSITIVE LOGITS
     Proust
    1.11
     Paddington
    0.94
     Madeleine
    0.91
     Comb
    0.81
     Swann
    0.80
    Comb
    0.74
     tavola
    0.74
     establecidas
    0.72
     aislado
    0.72
    馐
    0.71
    Activations Density 0.016%

    No Known Activations