© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 28345
    Prev
    Next
    INDEX
    Explanations

    identifying locations

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

    The neuron strongly activates on punctuation tokens—especially commas (and to a lesser extent periods)—throughout the text.

    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
     then
    -1.38
     оружи
    -1.13
     those
    -1.13
     ultimately
    -1.12
     OGSÅ
    -1.09
     example
    -1.08
     data
    -1.05
    deutung
    -1.05
     while
    -1.05
     increase
    -1.05
    POSITIVE LOGITS
     is
    1.15
     surrounded
    1.09
     location
    1.01
    ใกล้
    1.00
     hóspedes
    0.98
     where
    0.97
     comprends
    0.97
    进了
    0.96
    ギリ
    0.95
    жной
    0.94
    Activations Density 0.086%

    No Known Activations