© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 2818
    Prev
    Next
    INDEX
    Explanations

    words which are used positively and which seem relatively sophisticated and educated.

    oai_token-act-pair · gemini-2.0-flash

    Code variables

    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
    Negative Logits
     lèvres
    -1.15
     vœux
    -1.11
     épaules
    -1.05
     prêtres
    -1.05
     ouvriers
    -1.01
     négociations
    -0.99
     cérami
    -0.99
     dépens
    -0.98
     pouvoirs
    -0.96
     prisonniers
    -0.94
    POSITIVE LOGITS
     outlets
    0.65
     initComponents
    0.64
     examples
    0.63
     ones
    0.63
     assets
    0.63
     sources
    0.62
     constants
    0.62
     tokens
    0.60
     targets
    0.59
     sights
    0.59
    Activations Density 6.458%

    No Known Activations