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. 445
    Prev
    Next
    INDEX
    Explanations

    attends to the opening and closing statements within the legal context from blank tokens

    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
    0:0.00
    1:0.02
    2:0.06
    3:0.83
    4:0.01
    5:0.02
    6:0.02
    7:0.01
    Negative Logits
    ↵↵
    -1.32
    <eos>
    -0.83
    ↵
    -0.59
     
    -0.47
    ↵↵↵
    -0.44
     The
    -0.40
     (
    -0.39
     In
    -0.39
     A
    -0.38
     "
    -0.36
    POSITIVE LOGITS
     itſelf
    0.83
     ―――――
    0.72
     myſelf
    0.71
     Efq
    0.71
     iſt
    0.69
     Anſ
    0.69
     تضيفلها
    0.69
     themſelves
    0.68
     pleaſure
    0.68
     виправивши
    0.67
    Activations Density 1.768%

    No Known Activations