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. 18-GEMMASCOPE-RES-16K
    6. 592
    Prev
    Next
    INDEX
    Explanations

    references to different Pacific regions and ethnicities

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

    Following a body of water, describes peoples/ethnicities

    np_acts-logits-general · gemini-2.0-flash

    Pacific regions and countries

    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
     للاسماء
    -0.75
    MigrationBuilder
    -0.65
    AccessorTable
    -0.62
    rhosis
    -0.56
     engraçadas
    -0.55
    reportWebVitals
    -0.53
    MessageTagHelper
    -0.51
     دیکھیے
    -0.49
    ableView
    -0.49
    phology
    -0.49
    POSITIVE LOGITS
     region
    1.03
     countries
    0.96
    region
    0.88
     Region
    0.86
     Countries
    0.79
     nations
    0.76
     regions
    0.76
    Region
    0.76
     COUNTRIES
    0.69
    countries
    0.69
    Activations Density 0.210%

    No Known Activations