© 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. 22-GEMMASCOPE-RES-131K
    7. 34367
    Prev
    Next
    INDEX
    Explanations

    Extracellular vascular five

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

    structured documentation text, particularly in HTML, XML, or markup languages.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on numeric tokens (numbers) in the text.

    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
     brisa
    -1.18
     cera
    -1.07
    Ꭽ
    -1.07
     bricolage
    -1.07
     dingin
    -1.06
     regata
    -1.06
    ஜ
    -1.05
     langit
    -1.05
     sku
    -1.04
    军的
    -1.03
    POSITIVE LOGITS
     {
    1.05
    どちら
    1.02
    ἵ
    1.00
    氾
    0.99
    眺望
    0.98
    协助
    0.97
    stasy
    0.94
     deft
    0.94
    ppar
    0.91
     منطقة
    0.91
    Activations Density 0.002%

    No Known Activations