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. 5659
    Prev
    Next
    INDEX
    Explanations

    attends to repetition of certain terms or phrases from tokens that refer back to prior occurrences in the text

    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
    0:0.08
    1:0.09
    2:0.10
    3:0.26
    4:0.26
    5:0.02
    6:0.09
    7:0.07
    Negative Logits
    hast
    -0.23
    dom
    -0.22
    Sidenav
    -0.22
    be
    -0.22
     har
    -0.22
    IM
    -0.21
    GeneratedMessage
    -0.21
    ges
    -0.21
    mp
    -0.21
    ass
    -0.21
    POSITIVE LOGITS
     kasarigan
    0.39
     незавершена
    0.38
    AddTagHelper
    0.38
     defaultstate
    0.37
     jScrollPane
    0.35
     privées
    0.35
     للمعارف
    0.35
    InjectAttribute
    0.33
    WriteTagHelper
    0.32
    complexType
    0.32
    Activations Density 0.237%

    No Known Activations