© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 79796
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "world."

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

    world and its global equivalents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    <%@
    -0.44
     sla
    -0.42
    Fernseh
    -0.41
     deposit
    -0.41
     PROCEDURES
    -0.41
     lazy
    -0.40
     procedure
    -0.39
    Tracey
    -0.39
     Lazy
    -0.39
    ילום
    -0.39
    POSITIVE LOGITS
     mundo
    0.86
     world
    0.86
    world
    0.85
     wereld
    0.76
     WORLD
    0.75
    World
    0.72
    โลก
    0.71
     mondo
    0.70
     świecie
    0.70
    WORLD
    0.69
    Activations Density 0.026%

    No Known Activations