© 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. 39844
    Prev
    Next
    INDEX
    Explanations

    json serialization tags

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

    The neuron fires on numeric literals (digit sequences) in code, especially the numbers in struct‐tag annotations (e.g. those three‐digit values after `json:`).

    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
     («
    -1.46
    áver
    -1.43
     ovat
    -1.42
     caract
    -1.38
     condam
    -1.36
     (‘
    -1.36
    ꦢ
    -1.35
     environmentally
    -1.33
     Thermodynamics
    -1.33
     wrought
    -1.32
    POSITIVE LOGITS
     importance
    1.45
    cún
    1.43
     this
    1.41
     cocción
    1.40
    preneurs
    1.30
    龶
    1.30
    </i>
    1.27
    Nobody
    1.24
    遑
    1.24
    andosi
    1.23
    Activations Density 0.004%

    No Known Activations