© 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. 10-GEMMASCOPE-RES-131K
    7. 3818
    Prev
    Next
    INDEX
    Explanations

    chess titles and pieces

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

    The neuron selectively activates on mentions of chess piece names and related chess roles (e.g. pawn, bishop, queen, master).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    (“
    -1.95
    atized
    -1.65
     („
    -1.59
    ologic
    -1.52
    atique
    -1.51
     apellidos
    -1.44
     véritable
    -1.43
     lacivert
    -1.42
    included
    -1.41
     RIV
    -1.41
    POSITIVE LOGITS
     we
    2.03
     our
    1.98
     After
    1.67
     With
    1.64
     If
    1.61
     GREAT
    1.48
     JUST
    1.48
     Since
    1.45
     Following
    1.41
     ONLY
    1.41
    Activations Density 0.004%

    No Known Activations