Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 5252
    Prev
    Next
    INDEX
    Explanations

    phrases related to uncertainties and questions about actions

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

    Japanese sentence ending particles/polite phrases

    np_acts-logits-general · gemini-2.0-flash

    Japanese sentence endings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    •expression
    korean invoking-maybe

    No Comments

    Negative Logits
    >");
    
    -0.69
    %");
    -0.62
    "]);
    
    -0.62
     "");
    
    -0.62
    AutoField
    -0.60
    kloped
    -0.59
     />";
    -0.59
    )");
    
    -0.58
    ьаж
    -0.58
     _)
    -0.57
    POSITIVE LOGITS
     typelib
    0.61
    providedIn
    0.59
    ResponseWriter
    0.59
     المعيارى
    0.57
    AutoresizingMask
    0.57
     виправивши
    0.56
    DoubleQuotes
    0.56
     препратки
    0.53
    formik
    0.51
     Walkover
    0.50
    Activations Density 0.033%

    No Known Activations