© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 39081
    Prev
    Next
    INDEX
    Explanations

    api calls and database region

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

    This neuron fires on XML/TS markup tags and structural elements (e.g. the `<context>`, `<message>`, and related tag boundaries) in translation files.

    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
    Cinnamon
    -0.84
    榛
    -0.84
     manif
    -0.83
     gehalten
    -0.82
     Histor
    -0.80
     niz
    -0.77
    жня
    -0.77
    ꯑ
    -0.76
    drücke
    -0.76
     each
    -0.76
    POSITIVE LOGITS
     find
    0.98
     then
    0.97
     chosen
    0.96
     found
    0.92
     required
    0.91
     subsequently
    0.87
    选择
    0.87
     before
    0.86
     preliminary
    0.85
     obtain
    0.84
    Activations Density 0.012%

    No Known Activations