© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 10923
    Prev
    Next
    INDEX
    Explanations

    words related to business goals, numbers, and percentages

    oai_token-act-pair · gemini-2.0-flash

    variations of the word "break"

    np_token-act-pair-logits · gpt-4o-mini

    break

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
    /*
    -0.54
     Boas
    -0.54
    setCellValue
    -0.52
    kuuta
    -0.51
     fusi
    -0.50
    Alike
    -0.49
    jamientos
    -0.49
    MessageState
    -0.48
     ganda
    -0.48
    OGND
    -0.47
    POSITIVE LOGITS
     break
    1.77
     Break
    1.61
    break
    1.54
    Break
    1.43
     breaks
    1.39
     BREAK
    1.38
     Breaks
    1.31
     breaking
    1.30
     bre
    1.30
     broke
    1.29
    Activations Density 1.351%

    No Known Activations