© 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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 4289
    Prev
    Next
    INDEX
    Explanations

    conditions related to the state and actions of threads in a programming context

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

    Code, parentheses, or function-related syntax

    np_acts-logits-general · gemini-2.0-flash

    closing parenthesis or brace

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    клопе
    -0.89
    .";
    
    -0.85
    .)}
    -0.79
    _;
    
    -0.77
     ujednoznacz
    -0.76
    ";}
    -0.75
    ;}
    
    -0.74
    ."));
    -0.73
    .");
    
    -0.73
    '],
    
    -0.72
    POSITIVE LOGITS
    )
    0.46
     préfère
    0.46
     {
    0.43
     veuillez
    0.43
     accepté
    0.42
     weet
    0.41
     فإن
    0.40
     pihaknya
    0.40
     we
    0.40
    ↵↵
    0.40
    Activations Density 0.096%

    No Known Activations