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. 6546
    Prev
    Next
    INDEX
    Explanations

    attends to code blocks or functions indicated by curly braces from corresponding closing curly braces as endpoints

    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
    1:0.15
    2:0.35
    3:0.08
    4:0.08
    5:0.06
    6:0.05
    7:0.08
    Negative Logits
    ikh
    -0.32
    RepeatedField
    -0.28
     هستیم
    -0.27
    ambar
    -0.26
     your
    -0.26
    ts
    -0.26
    DAOImpl
    -0.26
     impo
    -0.26
     Sah
    -0.25
     comprehensively
    -0.25
    POSITIVE LOGITS
     Audiodateien
    0.52
    ValueStyle
    0.46
    namentales
    0.46
    DockStyle
    0.41
     незавершена
    0.40
    InvalidProtocol
    0.39
    invokeLater
    0.38
    erialization
    0.37
    contentLoaded
    0.37
     wireType
    0.37
    Activations Density 0.091%

    No Known Activations