© 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. 25704
    Prev
    Next
    INDEX
    Explanations

    winning money and jackpots

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

    The neuron activates primarily on numeric tokens or number-like sequences (e.g., digits, numeric expressions, dates, scores).

    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
    rophi
    -1.17
    ネジ
    -1.12
     its
    -1.12
    atique
    -1.06
    犷
    -1.00
    labeled
    -0.99
    offering
    -0.94
    AZIONE
    -0.94
    ference
    -0.94
     is
    -0.94
    POSITIVE LOGITS
     lucky
    1.41
     jackpot
    1.34
     luck
    1.28
    最初に
    1.16
    ambig
    1.10
     ratusan
    1.09
     göm
    1.09
     fortune
    1.09
     seguía
    1.08
    Only
    1.04
    Activations Density 0.029%

    No Known Activations