© 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. 84595
    Prev
    Next
    INDEX
    Explanations

    battleships and cruisers

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

    The neuron fires on mentions of naval warships (types like “auxiliary cruiser,” “battleship,” “cruiser”) and their related operations or actions (captured, supplied, separated).

    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
     Trudeau
    -0.86
     Frisch
    -0.80
     SEASON
    -0.80
     Schweizer
    -0.77
     Genève
    -0.76
    =]
    -0.74
     天然
    -0.74
    bpm
    -0.73
     laboral
    -0.73
    FUENTES
    -0.72
    POSITIVE LOGITS
     battleship
    2.00
     Battles
    1.38
    Battles
    1.19
     ships
    1.10
     battles
    1.06
    battles
    1.03
     лин
    1.02
     кора
    0.99
     ship
    0.98
     cruiser
    0.97
    Activations Density 0.031%

    No Known Activations