© 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. 7000
    Prev
    Next
    INDEX
    Explanations

    start and end of sequences

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

    The main thing this neuron does is detect numeric statistics—scores, counts, and other numbers—especially in sports‐style recaps.

    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
     первого
    -1.16
    First
    -1.11
    first
    -1.07
     первом
    -1.01
     First
    -0.98
     recentemente
    -0.95
     prequel
    -0.94
     récemment
    -0.93
     pierwszy
    -0.91
     первый
    -0.88
    POSITIVE LOGITS
     start
    1.37
     starts
    1.36
     Beginn
    1.20
     third
    1.18
     begins
    1.13
     started
    1.11
     begin
    1.04
     in
    1.04
    第三
    0.97
    เริ่ม
    0.95
    Activations Density 0.011%

    No Known Activations