© 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. 19-GEMMASCOPE-TRANSCODER-16K
    7. 1272
    Prev
    Next
    INDEX
    Explanations

    a grab bag of numerical and financial information, as well as some location names

    oai_token-act-pair · gemini-2.0-flash

    word fragments

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_19/width_16k/average_l0_12
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.19.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
     y
    -0.49
     [
    -0.48
     par
    -0.47
     ponte
    -0.46
     N
    -0.46
     L
    -0.45
     H
    -0.45
     p
    -0.45
    InitStruct
    -0.45
     P
    -0.44
    POSITIVE LOGITS
     Monfieur
    1.06
     Anſ
    1.03
    ſelves
    1.02
     itſelf
    1.02
     Efq
    1.01
     myſelf
    0.99
     ſtate
    0.98
    ſelf
    0.98
     ſta
    0.94
     Diſ
    0.93
    Activations Density 2.106%

    No Known Activations