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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 30348
    Prev
    Next
    INDEX
    Explanations

    references to the region of the South

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

    south direction and regions

    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
     wią
    -0.47
    abler
    -0.47
    BeginContext
    -0.47
     bagels
    -0.46
     fetus
    -0.45
     verknüp
    -0.45
     (;;)
    -0.45
    ejs
    -0.45
     isEnabled
    -0.44
    warran
    -0.44
    POSITIVE LOGITS
     South
    1.21
    South
    1.10
     SOUTH
    1.09
     south
    0.97
    south
    0.94
    SOUTH
    0.93
     Selatan
    0.85
     Southgate
    0.85
     North
    0.81
    North
    0.71
    Activations Density 0.018%

    No Known Activations