© 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. 6-GEMMASCOPE-TRANSCODER-16K
    7. 13965
    Prev
    Next
    INDEX
    Explanations

    years containing "18" or "19", often "2018" or "2019", as well as detecting some company names.

    oai_token-act-pair · gemini-2.0-flash

    Years with 8 or 9

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_6/width_16k/average_l0_95
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.6.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
    \{\\
    -0.79
     ExecuteAsync
    -0.67
    Rüyada
    -0.60
    DockStyle
    -0.60
    AnchorStyles
    -0.59
    DoubleQuotes
    -0.55
    SqlDataReader
    -0.53
    SharedDtor
    -0.53
    itzenden
    -0.52
    LookAnd
    -0.52
    POSITIVE LOGITS
    ########.
    0.52
    OGND
    0.50
    ătoare
    0.50
    <bos>
    0.49
    monių
    0.47
    ⓧ
    0.44
     fjor
    0.43
    jecha
    0.43
     홍
    0.43
     humaines
    0.42
    Activations Density 2.969%

    No Known Activations