© 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. 90801
    Prev
    Next
    INDEX
    Explanations

    buffer flushing

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

    The neuron activates on numeric literals (sequences of digits) in code.

    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
    fillText
    -0.80
     yaitu
    -0.77
    setter
    -0.72
    OMG
    -0.72
     wrze
    -0.71
     invece
    -0.71
    Saturday
    -0.70
    isteria
    -0.69
    цю
    -0.69
    holi
    -0.69
    POSITIVE LOGITS
    fflush
    2.63
     flush
    1.98
    flush
    1.84
     Flush
    1.61
     flushing
    1.55
    Flush
    1.55
    FLUSH
    1.50
     flushed
    1.40
     Flushing
    1.35
     rinsing
    0.96
    Activations Density 0.011%

    No Known Activations