© 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. 112740
    Prev
    Next
    INDEX
    Explanations

    past participle phrases

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

    The neuron activates on numeric expressions—tokens representing quantities, counts, dates, or statistical figures.

    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
      
    -1.58
    </strong>
    -1.27
    嫌弃
    -1.23
     guerrero
    -1.22
    again
    -1.20
    她的
    -1.20
    vertebrates
    -1.20
    with
    -1.19
    two
    -1.19
    其他的
    -1.17
    POSITIVE LOGITS
     expliqué
    1.63
    ")).
    1.45
     Você
    1.41
     seperti
    1.36
     esimerkiksi
    1.34
    🛇
    1.34
     possesso
    1.29
     verbess
    1.29
     demandé
    1.28
    >")
    1.27
    Activations Density 0.042%

    No Known Activations