© 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. 34-GEMMASCOPE-RES-131K
    7. 63496
    Prev
    Next
    INDEX
    Explanations

    code coverage and publishing

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

    The neuron is keyed to dotted namespace or package‐coordinate tokens (e.g. the segments and dots in Maven/GAV or Java package names).

    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
    пор
    -0.78
    畴
    -0.78
     mewakili
    -0.77
     YAML
    -0.77
    Inset
    -0.75
     backbone
    -0.71
    мяг
    -0.70
     yellowish
    -0.70
    nissen
    -0.69
     لب
    -0.69
    POSITIVE LOGITS
     coverage
    1.63
    coverage
    1.41
    Coverage
    1.20
     Coverage
    1.13
     publish
    1.08
     cobertura
    1.01
    cov
    1.00
    COVER
    0.99
     COVERAGE
    0.98
    cover
    0.93
    Activations Density 0.096%

    No Known Activations