© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 4314
    Prev
    Next
    INDEX
    Explanations

    words and phrases indicating that something is composed of multiple sub-elements

    oai_token-act-pair · gemini-2.0-flash

    composition

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
    featureID
    -0.76
    windowFixed
    -0.76
    IntoConstraints
    -0.70
    +#+#
    -0.68
    CppMethod
    -0.66
     ProtoMessage
    -0.66
     nakalista
    -0.66
    jsxFileName
    -0.66
    RegressionTest
    -0.65
    setVerticalGroup
    -0.64
    POSITIVE LOGITS
     made
    2.22
    made
    2.03
     Made
    1.95
    Made
    1.90
     composed
    1.77
     MADE
    1.74
    composed
    1.54
     Composed
    1.52
    MADE
    1.50
     comprised
    1.31
    Activations Density 2.703%

    No Known Activations