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. 10-GEMMASCOPE-ATT-16K
    6. 5818
    Prev
    Next
    INDEX
    Explanations

    attends to specific verbs that indicate speaking or stating from conjunctions that connect actions or statements

    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
    0:0.13
    1:0.15
    2:0.12
    3:0.10
    4:0.09
    5:0.06
    6:0.12
    7:0.20
    Negative Logits
    INSEE
    -0.28
    ::*;
    -0.28
    Élet
    -0.25
     Ballad
    -0.25
     apre
    -0.25
    _{+}
    -0.24
    డా
    -0.24
    ])){
    -0.24
    ])):
    -0.24
     examen
    -0.24
    POSITIVE LOGITS
     незавершена
    0.31
    TintMode
    0.31
    protoimpl
    0.27
    whole
    0.26
     Whole
    0.25
     initComponents
    0.24
    脚注の使い方
    0.24
    IsContent
    0.24
    Whole
    0.23
    Glit
    0.23
    Activations Density 0.588%

    No Known Activations