© 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. 22-GEMMASCOPE-RES-131K
    7. 70717
    Prev
    Next
    INDEX
    Explanations

    trouble and question marks

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

    The neuron reliably fires strongest on the final token of each snippet, effectively marking the end of a text block.

    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
     which
    -1.07
     univerz
    -1.06
     pejabat
    -1.02
    cobacterium
    -1.01
    جري
    -1.00
     because
    -0.94
    ıldığı
    -0.93
    ≦
    -0.93
     also
    -0.93
     Meille
    -0.93
    POSITIVE LOGITS
     rembour
    1.02
    including
    1.00
    則是
    0.98
     ज्ञात
    0.98
    ולות
    0.96
    רוק
    0.96
    ופר
    0.95
    רוש
    0.93
    frequently
    0.91
     bloem
    0.91
    Activations Density 0.035%

    No Known Activations