© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 84498
    Prev
    Next
    INDEX
    Explanations

    unicorn value

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

    The neuron strongly activates on numerical tokens—especially monetary amounts, percentages, and other numeric expressions.

    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
    TabItem
    -0.73
    firmware
    -0.70
     жов
    -0.69
     Vorles
    -0.68
     rela
    -0.68
    browserify
    -0.67
    vetica
    -0.66
     specify
    -0.66
     porcentual
    -0.65
     straightforward
    -0.65
    POSITIVE LOGITS
     unicorn
    1.77
     unicorns
    1.50
     Unicorn
    1.49
    unicorn
    1.31
    Unicorn
    1.27
     Valuation
    1.10
     valuation
    1.09
     unicornio
    1.02
    Soft
    0.98
    🦄
    0.95
    Activations Density 0.017%

    No Known Activations