© 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. 25-GEMMASCOPE-TRANSCODER-16K
    7. 7387
    Prev
    Next
    INDEX
    Explanations

    time, date, and number information within a structured log format, as well as math forumulas. I also noticed 'Dar' sometimes.

    oai_token-act-pair · gemini-2.0-flash

    0

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.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
     PT
    -1.35
     BT
    -1.30
     CT
    -1.29
     RT
    -1.28
     FT
    -1.25
     MT
    -1.24
     CST
    -1.20
     HT
    -1.17
     DT
    -1.17
     NT
    -1.16
    POSITIVE LOGITS
    uarga
    0.44
    胧
    0.42
    UrlResolution
    0.40
    archiviato
    0.40
    liefer
    0.39
    CellStyle
    0.38
     EClass
    0.38
    änä
    0.38
    filer
    0.38
    MeasureSpec
    0.38
    Activations Density 1.885%

    No Known Activations