© 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. 60114
    Prev
    Next
    INDEX
    Explanations

    orders, sales, purchases, receipts

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

    The neuron detects code‐style identifier tokens—alphanumeric strings with uppercase letters, digits, underscores, or CamelCase that are typical of programming symbols.

    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
    parseLong
    -0.79
     obtainable
    -0.77
    льное
    -0.77
     Geschäft
    -0.75
    狰
    -0.73
    yya
    -0.73
     dostup
    -0.72
    crack
    -0.72
    jutkan
    -0.72
    WALL
    -0.71
    POSITIVE LOGITS
     credit
    1.10
     orders
    1.03
     order
    1.02
     return
    0.97
     Order
    0.96
     note
    0.94
     bill
    0.93
     purchase
    0.92
    Order
    0.92
    Credit
    0.92
    Activations Density 0.015%

    No Known Activations