© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 128367
    Prev
    Next
    INDEX
    Explanations

    need or looking for

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

    The neuron fires on fleeting promotional or testimonial turns of phrase expressing a need or suitability—e.g. “just what you need,” “what I need to,” or “what X needs.”

    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
     will
    -1.31
    …(
    -1.21
     один
    -1.18
    をクリック
    -1.18
     одна
    -1.16
    而成
    -1.16
    : 
    -1.15
    foundland
    -1.14
    をご覧ください
    -1.14
    なー
    -1.13
    POSITIVE LOGITS
     moeten
    1.52
    
    1.38
    }'.
    1.32
    鲅
    1.32
    tgärder
    1.31
    还是
    1.30
     reyes
    1.30
    薙
    1.29
     zwierząt
    1.28
    듭
    1.27
    Activations Density 0.031%

    No Known Activations