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. 5616
    Prev
    Next
    INDEX
    Explanations

    attends to specific punctuation marks and related content from unrelated tokens

    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
    0:0.08
    1:0.11
    2:0.10
    3:0.09
    4:0.08
    5:0.11
    6:0.26
    7:0.12
    Negative Logits
    GEBURTS
    -0.39
    Vidite
    -0.37
     AspNetCore
    -0.35
     >=",
    -0.34
     <>",
    -0.34
    杵
    -0.33
    httphttps
    -0.33
    odils
    -0.33
    DAOImpl
    -0.32
     AttributeSet
    -0.31
    POSITIVE LOGITS
     незавершена
    0.29
    Activités
    0.29
    WriteLiteral
    0.28
    urlpatterns
    0.26
     satunya
    0.26
     المعيارى
    0.25
     anledning
    0.25
     grunn
    0.24
     dasarnya
    0.24
     "
    0.23
    Activations Density 0.000%

    No Known Activations