© 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. 81308
    Prev
    Next
    INDEX
    Explanations

    collectible figurines

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

    The neuron consistently responds to tokens that are part of product‐style headings or codes—i.e. title-case words and numeric strings (years, model numbers, version numbers).

    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
    VGND
    -0.88
     experimental
    -0.84
     wiki
    -0.78
    实验
    -0.75
    álás
    -0.73
    legd
    -0.72
    autoreleasepool
    -0.72
     γ
    -0.72
    観光
    -0.72
    DIRECTORY
    -0.72
    POSITIVE LOGITS
     figurines
    1.15
     figurine
    1.13
     porcelain
    1.09
     COLLECT
    1.06
    HAEL
    1.05
     Figur
    1.04
     collectible
    0.98
    Collector
    0.96
     collectors
    0.94
     Collector
    0.93
    Activations Density 0.031%

    No Known Activations