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. Attention Out - 16k
    5. 10-GEMMASCOPE-ATT-16K
    6. 5957
    Prev
    Next
    INDEX
    Explanations

    attends to style-related tokens marked in code from adjacent key extractor tokens related to lists

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.attn.hook_z
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.10
    1:0.14
    2:0.07
    3:0.05
    4:0.15
    5:0.30
    6:0.04
    7:0.11
    Negative Logits
    queous
    -0.31
    ukone
    -0.31
     parch
    -0.31
    iconque
    -0.31
     RIPRODUZIONE
    -0.30
     Waterhouse
    -0.30
    hopped
    -0.30
     objectively
    -0.30
     snowflakes
    -0.29
    מבר
    -0.29
    POSITIVE LOGITS
     الاطلاع
    0.37
    ReusableCell
    0.35
    SuppressMessage
    0.34
    twimg
    0.33
    MemoryWarning
    0.33
    InjectMocks
    0.33
    ضو
    0.32
    LayoutInflater
    0.32
     char
    0.32
    TagMode
    0.32
    Activations Density 0.001%

    No Known Activations