© 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. 43939
    Prev
    Next
    INDEX
    Explanations

    proof, investigation, showing results

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

    The neuron is specialized for detecting numeric expressions (both integers and decimal values) in the document.

    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
     it
    -2.08
     each
    -1.23
     uses
    -1.17
    éné
    -1.13
     inclui
    -1.12
     that
    -1.10
     use
    -1.09
     at
    -1.07
     them
    -1.07
     fordert
    -1.06
    POSITIVE LOGITS
    ům
    1.41
     вещества
    1.22
     Temmuz
    1.20
    (),
    1.18
     listopada
    1.13
    ordes
    1.11
    ligheter
    1.10
     starten
    1.09
     向け
    1.08
    losen
    1.07
    Activations Density 0.216%

    No Known Activations