© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 4588
    Prev
    Next
    INDEX
    Explanations

    seemingly random words, sometimes near the word "and", and sometimes near other formatting elements

    oai_token-act-pair · gemini-2.0-flash

    and, Y

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
    and
    -0.82
    ernalia
    -0.75
     avoient
    -0.72
     étoient
    -0.69
     noires
    -0.69
     Andra
    -0.66
     Armas
    -0.66
     Steck
    -0.66
    AND
    -0.65
     loto
    -0.64
    POSITIVE LOGITS
     y
    1.31
     Y
    1.14
     ya
    0.70
     propOrder
    0.70
     autorytatywna
    0.69
    ftagPool
    0.68
     or
    0.68
    ImageContext
    0.67
     og
    0.66
    fjspx
    0.66
    Activations Density 2.395%

    No Known Activations