© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2539
    Prev
    Next
    INDEX
    Explanations

    the presence of error handling constructs in code, particularly those related to "try" and "catch" statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    programming code structures, specifically try-catch blocks and method definitions.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     мәкал
    -0.95
     ویکی‌پدی
    -0.93
     transfieras
    -0.92
     незавершена
    -0.92
    ьаж
    -0.91
     zwiſchen
    -0.90
    ésultats
    -0.90
    AndEndTag
    -0.85
    <unused3>
    -0.84
    <unused41>
    -0.84
    POSITIVE LOGITS
    try
    0.48
    VStack
    0.40
     there
    0.35
    ,
    0.35
     There
    0.33
    :
    0.32
    There
    0.32
    ↵
    0.32
    Sub
    0.31
    ;
    0.30
    Activations Density 0.095%

    No Known Activations