© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 98167
    Prev
    Next
    INDEX
    Explanations

    code imports and templates

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

    The neuron strongly fires on closing‐block tokens—e.g. HTML closing tags (“</…>”) and code block closing braces (“}”)—marking the ends of structured regions.

    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
     різних
    -1.14
     différentes
    -1.13
     défini
    -1.13
    现在的
    -1.10
     mauva
    -1.08
     verschillende
    -1.05
     désormais
    -1.04
    鼉
    -1.01
     実例
    -1.01
     konsult
    -1.00
    POSITIVE LOGITS
     and
    1.23
    ,
    1.05
     in
    1.03
     Remember
    1.03
     Before
    1.02
     on
    1.02
     should
    1.00
     IN
    0.98
    .
    0.96
    ↵↵↵↵↵↵↵
    0.93
    Activations Density 0.001%

    No Known Activations