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. 6023
    Prev
    Next
    INDEX
    Explanations

    attends to the token "het" from tokens that provide additional context or specification

    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
    0:0.08
    1:0.03
    2:0.05
    3:0.05
    4:0.10
    5:0.45
    6:0.12
    7:0.08
    Negative Logits
    OrEqualTo
    -0.28
     the
    -0.26
    LEncoder
    -0.25
    BeginContext
    -0.25
    ľov
    -0.24
    ̲
    -0.24
    GLE
    -0.24
    azgo
    -0.24
    fores
    -0.24
     CreateTagHelper
    -0.24
    POSITIVE LOGITS
    ########.
    0.45
    __':
    
    0.41
    expandindo
    0.40
    __":
    
    0.40
    Према
    0.38
     kasarigan
    0.36
    CppCodeGen
    0.36
    iecie
    0.36
    AsUp
    0.35
     المعيارى
    0.35
    Activations Density 0.033%

    No Known Activations