© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 13311
    Prev
    Next
    INDEX
    Explanations

    meeting locations and times

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

    The neuron detects tokens that appear in scheduling and location phrases—prepositions like “on” and “in” alongside numbers and ordinals that specify dates, times, or places.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     {
    -2.36
     you
    -2.11
        
    -1.98
            
    -1.91
    <i>
    -1.84
     of
    -1.79
    }
    -1.79
     quando
    -1.77
     ajoute
    -1.73
         
    -1.72
    POSITIVE LOGITS
    嗞
    2.14
    ꧁
    2.11
     malheure
    2.08
    kemeja
    2.02
    勠
    1.98
    歃
    1.98
     froide
    1.87
    ཫ
    1.79
    ㌘
    1.78
    
    1.78
    Activations Density 0.019%

    No Known Activations