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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 56445
    Prev
    Next
    INDEX
    Explanations

    references to specific locations, particularly in Queensland, Australia

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

    Queensland places

    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
     gynhyrchwyd
    -0.47
     Korean
    -0.46
    ussis
    -0.46
     ब्रेकडाउन
    -0.45
    ekwondo
    -0.44
    참고
    -0.44
    出版年
    -0.43
    AsUp
    -0.42
     Yakima
    -0.41
     Tae
    -0.41
    POSITIVE LOGITS
    Queensland
    0.80
     Queensland
    0.79
    Brisbane
    0.74
     Brisbane
    0.73
    Florida
    0.68
    mybatisplus
    0.61
     QLD
    0.61
     pinulongan
    0.60
     Florida
    0.60
     tropicales
    0.60
    Activations Density 0.007%

    No Known Activations