© 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-2-2B
    5. 24-GEMMASCOPE-TRANSCODER-16K
    7. 11570
    Prev
    Next
    INDEX
    Explanations

    This neuron appears to be strongly activated by German language, especially grammatical structures, verbs, and sentence construction.

    oai_token-act-pair · gemini-2.0-flash

    commas

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ')):
    -0.51
    ╝
    -0.46
    --)
    -0.46
    的是
    -0.46
     rằng
    -0.43
    ']):
    -0.42
    seitige
    -0.42
    Enllaces
    -0.41
    ChildScrollView
    -0.41
    неопр
    -0.41
    POSITIVE LOGITS
    ,
    2.11
     ,
    0.85
    ,<
    0.80
    ,
    0.70
    ,{\
    0.69
    ，
    0.69
    ,\
    0.69
    ،
    0.67
    ,[
    0.65
    ,&
    0.64
    Activations Density 0.869%

    No Known Activations