© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 102566
    Prev
    Next
    INDEX
    Explanations

    occlusion of

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

    The neuron specializes in detecting the word “occlusion” (including its subparts like “clusion” or “oc-”) in 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_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
    ⸰
    -2.78
    埽
    -2.77
    蠓
    -2.72
    逦
    -2.64
     publice
    -2.52
     イラストレーション
    -2.50
    鷽
    -2.50
    
    -2.50
    齶
    -2.50
    CHREIB
    -2.45
    POSITIVE LOGITS
    ↵
    4.59
    </em>
    3.78
     $
    3.23
    n
    2.94
    }{
    2.44
    \_
    2.36
    </h4>
    2.33
    l
    2.25
     }
    2.25
    m
    2.22
    Activations Density 0.013%

    No Known Activations