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
    4. Attention Out - 16k
    5. 0-GEMMASCOPE-ATT-16K
    6. 1213
    Prev
    Next
    INDEX
    Explanations

    attends to tokens that indicate additional information or context from subsequent tokens that also provide additional details

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-att/layer_0/width_16k/average_l0_61
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.attn.hook_z
    Hook Layer
    0
    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
    0:0.01
    1:0.00
    2:0.02
    3:0.01
    4:0.00
    5:0.01
    6:0.61
    7:0.10
    8:0.00
    9:0.01
    10:0.00
    11:0.03
    12:0.01
    13:0.08
    14:0.02
    15:0.01
    Negative Logits
     autorytatywna
    -0.99
     ویکی‌پدی
    -0.87
    uxxxx
    -0.75
     Италијани
    -0.73
     noDo
    -0.68
    IUrlHelper
    -0.68
     betweenstory
    -0.68
    ंदीखरीदारी
    -0.68
     Мексичка
    -0.65
     kaarangay
    -0.64
    POSITIVE LOGITS
    1
    0.27
    2
    0.23
    _
    0.23
     
    0.22
        
    0.22
       
    0.22
    ↵
    0.21
      
    0.20
    3
    0.20
    0
    0.20
    Activations Density 0.160%

    No Known Activations