© 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. 22-GEMMASCOPE-RES-131K
    4. 26393
    Prev
    Next
    INDEX
    Explanations

    neighborhoods, questions, plumbing

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

    This neuron fires on multi‐token named entities and proper names (like organization titles, place names, or domain names/URLs).

    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
     vif
    -1.17
     lel
    -1.17
    ಱ
    -1.13
    *',
    -1.11
     surat
    -1.10
     ▸
    -1.09
    WRITER
    -1.08
    panes
    -1.07
    derry
    -1.07
    dreaming
    -1.06
    POSITIVE LOGITS
    最も
    1.23
     to
    1.05
    </h2>
    1.02
    }
    
    1.00
    </h4>
    0.98
    </b>
    0.98
    </strong>
    0.98
     має
    0.96
    最近
    0.96
    Have
    0.95
    Activations Density 0.001%

    No Known Activations