© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14531
    Prev
    Next
    INDEX
    Explanations

    type

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

    This neuron activates on programming identifiers or constants containing “TYPE” (e.g. OBJECT_TYPE, FIELD_TYPES).

    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
     believes
    -1.99
    uttgart
    -1.75
    o
    -1.73
    鬭
    -1.70
    饜
    -1.62
     takes
    -1.60
     riuscito
    -1.60
     plucked
    -1.59
     took
    -1.59
    Mereka
    -1.59
    POSITIVE LOGITS
    鈊
    2.02
     podem
    1.95
    ––
    1.84
    舠
    1.81
    '
    1.78
     ayant
    1.75
     inteligentes
    1.73
     VERY
    1.73
    ⚌
    1.73
    —-
    1.72
    Activations Density 0.018%

    No Known Activations