© 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. 24-GEMMASCOPE-TRANSCODER-16K
    7. 9052
    Prev
    Next
    INDEX
    Explanations

    seemingly random numerical data or figures in the text, possibly pulling other, adjacent words

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
    AndEndTag
    -0.91
     propOrder
    -0.90
    AsUp
    -0.89
     مشين
    -0.81
     nahilalakip
    -0.75
    Билгалдахарш
    -0.71
    UnusedPrivate
    -0.71
     noDo
    -0.71
    存于互联网档案馆
    -0.70
    ագրություններ
    -0.70
    POSITIVE LOGITS
    1
    0.74
     
    0.60
    2
    0.59
    m
    0.58
    0
    0.55
    3
    0.54
    M
    0.53
     Hyman
    0.52
      
    0.51
     Tears
    0.49
    Activations Density 47.320%

    No Known Activations