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-2-2B
    4. MLP Out - 16k
    5. 1-GEMMASCOPE-MLP-16K
    6. 8735
    Prev
    Next
    INDEX
    Explanations

    references to various football (soccer) tournaments, specifically the World Cup and Gold Cup events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k
    Configuration
    google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.1.hook_mlp_out
    Hook Layer
    1
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cup
    -1.59
    Cup
    -1.24
     cup
    -0.94
     CUP
    -0.93
    cup
    -0.92
     Cups
    -0.79
    カップ
    -0.64
    OrNil
    -0.61
    CUP
    -0.60
     للاسماء
    -0.57
    POSITIVE LOGITS
    ########.
    0.88
    ActivityCompat
    0.66
    addGroup
    0.62
     exécu
    0.59
     onCreateView
    0.59
     CYTHON
    0.59
    ,:,
    0.56
     heureuse
    0.56
     محفوظة
    0.55
     étrang
    0.54
    Activations Density 0.211%

    No Known Activations