© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 39041
    Prev
    Next
    INDEX
    Explanations

    production and related samples

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

    The neuron primarily activates on numerical tokens—digits, decimal numbers, and counts.

    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
    omania
    -0.97
    Prie
    -0.92
     pengar
    -0.90
     munici
    -0.90
    belline
    -0.90
    nellement
    -0.89
     Alors
    -0.88
     ihop
    -0.87
     Suom
    -0.85
    lı
    -0.85
    POSITIVE LOGITS
     where
    1.09
     tipo
    1.02
     Diese
    0.99
     móg
    0.93
     donde
    0.91
     selaku
    0.89
     otros
    0.88
     weitere
    0.88
     their
    0.87
     weinig
    0.87
    Activations Density 0.142%

    No Known Activations