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. 6203
    Prev
    Next
    INDEX
    Explanations

    attends to certain table or model indicators from other similarly formatted table or model indicators

    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
    0:0.11
    1:0.13
    2:0.13
    3:0.08
    4:0.13
    5:0.05
    6:0.11
    7:0.22
    Negative Logits
    expandindo
    -0.56
    adpleegd
    -0.38
    BeginContext
    -0.36
    SharedDtor
    -0.36
    esez
    -0.35
     resourceCulture
    -0.34
    Personendaten
    -0.34
     Audiodateien
    -0.33
    
    -0.33
    ::~
    -0.32
    POSITIVE LOGITS
    crombie
    0.27
     Pov
    0.26
    Geographie
    0.24
    JAXB
    0.24
    Belf
    0.24
     Belf
    0.23
     afa
    0.23
    getAdapter
    0.22
     pov
    0.22
     therefrom
    0.21
    Activations Density 0.124%

    No Known Activations