Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 6587
    Prev
    Next
    INDEX
    Explanations

    attends to the "import" statement from related tokens representing designated components or modules

    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
    0:0.12
    1:0.24
    2:0.10
    3:0.10
    4:0.15
    5:0.10
    6:0.06
    7:0.08
    Negative Logits
    *********/
    -0.45
    IRP
    -0.36
     Racine
    -0.35
    loroethene
    -0.35
     Widodo
    -0.34
    רושלים
    -0.33
     Farah
    -0.33
     Brah
    -0.32
    ?>
    -0.32
     "..\..\
    -0.32
    POSITIVE LOGITS
    ########.
    0.52
     '\\;'
    0.47
    LookAnd
    0.47
    KURZBESCHREIBUNG
    0.44
     transférez
    0.44
    ServletConfig
    0.44
    SharedDtor
    0.43
    hilt
    0.42
     ProtoMessage
    0.41
    expandindo
    0.41
    Activations Density 0.004%

    No Known Activations