© 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. 34-GEMMASCOPE-RES-131K
    7. 78224
    Prev
    Next
    INDEX
    Explanations

    names and places that are

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

    The neuron activates on capitalized named entities (proper names of products, companies, places, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    usura
    -1.18
    Would
    -1.05
    olescence
    -1.01
    adalah
    -0.97
     Would
    -0.97
     reç
    -0.96
     nadat
    -0.93
    []=$
    -0.92
    $).
    -0.92
     lihat
    -0.90
    POSITIVE LOGITS
     also
    1.38
     currently
    1.24
     is
    1.16
    かし
    1.07
     celebración
    1.04
     principalmente
    0.90
     zejména
    0.90
     formerly
    0.88
     celebrating
    0.88
     predominantly
    0.88
    Activations Density 0.001%

    No Known Activations