© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 4-GEMMASCOPE-TRANSCODER-16K
    7. 401
    Prev
    Next
    INDEX
    Explanations

    words that can be considered action verbs, especially those ending in "-ing", "ed", or involving complex processes.

    oai_token-act-pair · gemini-2.0-flash

    involve

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     involve
    -1.16
     involves
    -1.14
     consists
    -1.00
     consiste
    -0.91
     melibatkan
    -0.90
     involving
    -0.88
     consist
    -0.87
     implique
    -0.79
     includes
    -0.78
    <bos>
    -0.78
    POSITIVE LOGITS
     viewDidLoad
    0.54
     FontWeight
    0.47
    XmlAccessType
    0.45
    createNewFile
    0.45
     jeruk
    0.45
    migrationBuilder
    0.43
     ModelExpression
    0.43
     BERNAMA
    0.43
    oneofs
    0.42
    ↵↵
    0.42
    Activations Density 8.922%

    No Known Activations