© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 14263
    Prev
    Next
    INDEX
    Explanations

    mostly junk, but in particular it detects acronyms and codes, especially in scientific notation, and maybe in business too.

    oai_token-act-pair · gemini-2.0-flash

    technical language

    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
     متعلقه
    -0.88
    SequentialGroup
    -0.76
     المعيارى
    -0.70
    +#+#
    -0.65
     >=",
    -0.64
    帖最后由
    -0.64
     disambiguazione
    -0.62
    PerformLayout
    -0.61
     CanadaChoose
    -0.61
     ligiloj
    -0.61
    POSITIVE LOGITS
     D
    1.25
     DP
    1.20
     d
    1.17
    DP
    1.16
     DD
    1.14
     DG
    1.11
     DT
    1.10
    getD
    1.09
     DS
    1.09
     DM
    1.08
    Activations Density 13.801%

    No Known Activations