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-2-2B
    4. Attention Out - 16k
    5. 10-GEMMASCOPE-ATT-16K
    6. 6410
    Prev
    Next
    INDEX
    Explanations

    attends to tokens related to conjunctions or lists from other tokens in the same categories or contexts

    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
    1:0.11
    2:0.04
    3:0.03
    4:0.08
    5:0.54
    6:0.03
    7:0.03
    Negative Logits
    CloseOperation
    -0.39
    bleau
    -0.36
    ógy
    -0.36
     betweenstory
    -0.35
    UpInside
    -0.35
    ddots
    -0.35
    äumt
    -0.35
    ezy
    -0.34
    satunya
    -0.34
    <h6>
    -0.34
    POSITIVE LOGITS
     vindt
    0.27
    جستارهای
    0.27
    Citation
    0.27
     Bell
    0.26
     еды
    0.26
    AccessorTable
    0.26
     Keller
    0.25
     gaat
    0.25
     уче
    0.25
    ார்
    0.25
    Activations Density 0.312%

    No Known Activations