© 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. 90914
    Prev
    Next
    INDEX
    Explanations

    charset utf8

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

    This neuron detects code‐file metadata and directives—things like license headers, include/pragma statements, and other uppercase identifiers and version numbers common in source‐code files.

    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
    萝
    -0.93
     Swanson
    -0.69
    ideae
    -0.68
     asynchronously
    -0.66
     Nvidia
    -0.66
     naruto
    -0.66
    Honest
    -0.66
    习
    -0.66
    courtesy
    -0.65
    erialize
    -0.65
    POSITIVE LOGITS
     collation
    0.94
     estacionamiento
    0.87
    }}^{\
    0.87
    arel
    0.83
     mycelium
    0.80
     relish
    0.80
    夫妇
    0.77
    imals
    0.75
     produit
    0.74
     Jut
    0.74
    Activations Density 0.019%

    No Known Activations