© 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. 22-GEMMASCOPE-RES-131K
    7. 25980
    Prev
    Next
    INDEX
    Explanations

    Residue analysis

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

    default values in programming code, particularly in configuration settings.

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

    The neuron fires on numeric literals (number tokens) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    荨
    -0.93
    ggable
    -0.91
     Hoog
    -0.90
    この
    -0.88
     maten
    -0.88
    ј
    -0.87
    QUEUE
    -0.87
    もう
    -0.86
    genza
    -0.84
    append
    -0.84
    POSITIVE LOGITS
     defaultValue
    1.20
    animo
    1.05
     sayfas
    1.04
     prosegu
    1.03
     núcleos
    1.02
     asimismo
    1.00
     null
    0.99
     default
    0.98
    dati
    0.94
    exercice
    0.94
    Activations Density 0.001%

    No Known Activations