© 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. 283
    Prev
    Next
    INDEX
    Explanations

    enderio, minecraft, caspase, rotten, Olivier

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

    programming language constructs and file paths in code.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron recognizes identifiers and keywords in Java package-import paths (i.e. the components of fully qualified package or import declarations).

    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
     Benchmark
    -0.81
    Benchmark
    -0.80
     thut
    -0.79
    acters
    -0.77
    leren
    -0.76
    ジェラ
    -0.73
    ➮
    -0.73
     benchmark
    -0.72
     Hawkeye
    -0.72
    AKAN
    -0.71
    POSITIVE LOGITS
    digd
    0.85
    AccessFile
    0.80
     Ophthalmology
    0.78
     BorderSide
    0.75
     crystals
    0.75
     Ago
    0.74
    ColumnName
    0.73
    Boys
    0.73
     brewing
    0.70
     stare
    0.70
    Activations Density 0.012%

    No Known Activations