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. 0-GEMMASCOPE-ATT-16K
    6. 6077
    Prev
    Next
    INDEX
    Explanations

    attends to various types of food from related concepts identified in the text

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_0/width_16k/average_l0_104
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.attn.hook_z
    Hook Layer
    0
    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
    0:0.79
    1:0.01
    2:0.01
    3:0.02
    4:0.09
    5:0.01
    6:0.01
    7:0.03
    Negative Logits
     EconPapers
    -0.32
    ns
    -0.27
    vič
    -0.25
    mb
    -0.25
    -
    -0.25
    glfw
    -0.25
     sitting
    -0.24
     sugar
    -0.24
     b
    -0.23
     I
    -0.23
    POSITIVE LOGITS
     NgModule
    0.42
    irited
    0.40
     referenties
    0.39
     ainfi
    0.39
    InputBorder
    0.39
     doPost
    0.39
     MaterialApp
    0.38
    PMailer
    0.38
    ThroughAttribute
    0.38
    saraba
    0.38
    Activations Density 0.018%

    No Known Activations