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

    Neuronpedia

    APICircuit 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. 0-GEMMASCOPE-ATT-16K
    6. 6025
    Prev
    Next
    INDEX
    Explanations

    attends to the specific term "ingredients" from the broader context of components or elements referenced later 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
    0:0.72
    1:0.01
    2:0.01
    3:0.03
    4:0.11
    5:0.03
    6:0.02
    7:0.03
    Negative Logits
     HasFactory
    -0.35
    PerformLayout
    -0.35
     ویکی‌پدیا
    -0.34
     MediatR
    -0.33
    новниш
    -0.32
     minori
    -0.29
    bootstrapcdn
    -0.29
    gya
    -0.28
    DockStyle
    -0.28
    ]")]
    -0.28
    POSITIVE LOGITS
    dicionado
    0.34
    віть
    0.31
    cotch
    0.30
    Spoljašnje
    0.28
    cherichia
    0.28
     rằng
    0.27
    τία
    0.27
     kwamba
    0.27
    vician
    0.27
    ICATED
    0.26
    Activations Density 0.029%

    No Known Activations