© 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
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 5503
    Prev
    Next
    INDEX
    Explanations

    words and phrases commonly associated with e-commerce and customer satisfaction.

    oai_token-act-pair · gemini-2.0-flash

    verbs associated with actions or processes

    np_token-act-pair-logits · gpt-4o-mini

    E-commerce and buying

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
    DockStyle
    -0.78
    Personendaten
    -0.77
     consultato
    -0.67
     AssemblyCulture
    -0.64
     Sotheby
    -0.62
     asymmetric
    -0.61
     referenties
    -0.61
    awaiter
    -0.59
    (@"%@",
    -0.57
    ISupport
    -0.56
    POSITIVE LOGITS
    AddTagHelper
    0.76
     make
    0.57
     made
    0.57
     doing
    0.56
    Hacer
    0.56
     utafitiHapana
    0.56
     conduct
    0.55
     badan
    0.52
    Conduct
    0.52
     makes
    0.52
    Activations Density 1.758%

    No Known Activations