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. 5785
    Prev
    Next
    INDEX
    Explanations

    attends to the token "id" from tokens marked with "id" that appear later in the sequences

    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
    0:0.06
    1:0.14
    2:0.06
    3:0.04
    4:0.13
    5:0.43
    6:0.04
    7:0.06
    Negative Logits
     glo
    -0.20
    RenderAtEndOf
    -0.20
    <bos>
    -0.20
    ModelAdmin
    -0.20
     sendiri
    -0.19
     SUDOC
    -0.19
     focus
    -0.18
    らう
    -0.17
     for
    -0.16
     blo
    -0.16
    POSITIVE LOGITS
     متعلقه
    0.46
     resourceCulture
    0.40
     MainAxisSize
    0.36
    setVerticalGroup
    0.34
    IntoConstraints
    0.34
    ChildIndex
    0.34
    ArgsConstructor
    0.33
     consultato
    0.33
     CommonModule
    0.33
     وتسجيلات
    0.32
    Activations Density 0.035%

    No Known Activations