© 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. 34-GEMMASCOPE-RES-131K
    7. 2723
    Prev
    Next
    INDEX
    Explanations

    dates and timestamps

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

    This neuron primarily responds to multi‐digit numeric sequences (e.g., version numbers, IDs, timestamps).

    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
     portray
    -0.79
     infatti
    -0.76
    üglich
    -0.75
     reass
    -0.75
     Keepers
    -0.73
    PRODUCTION
    -0.73
     recite
    -0.73
     Deus
    -0.71
    ELT
    -0.71
    fabricated
    -0.71
    POSITIVE LOGITS
    席
    0.78
    стую
    0.77
    披
    0.75
    혔
    0.74
     Tất
    0.73
     alguma
    0.71
     aktive
    0.70
     dej
    0.70
     redaktion
    0.70
    ¿
    0.69
    Activations Density 0.074%

    No Known Activations