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. 5695
    Prev
    Next
    INDEX
    Explanations

    attends to numerical tokens denoting counts or quantities from earlier tokens that also indicate similar counts or quantities

    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
    0:0.04
    1:0.11
    2:0.06
    3:0.03
    4:0.17
    5:0.50
    6:0.02
    7:0.04
    Negative Logits
     للمعارف
    -0.29
     MOND
    -0.27
    таратура
    -0.26
    Personensuche
    -0.26
    Chham
    -0.25
    InputTagHelper
    -0.25
     AssemblyTitle
    -0.25
    aarrggbb
    -0.25
    verwijspagina
    -0.25
     وتسجيلات
    -0.25
    POSITIVE LOGITS
    __":
    
    0.24
    DataMember
    0.23
    aufen
    0.22
    /*++
    0.22
    alloon
    0.22
     پذ
    0.21
     Salt
    0.21
    readystatechange
    0.21
     electrónicos
    0.21
    lieder
    0.20
    Activations Density 0.260%

    No Known Activations