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. 139
    Prev
    Next
    INDEX
    Explanations

    attends to opening curly braces from token pairs involving a following token, regardless of context

    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
    0:0.03
    1:0.05
    2:0.05
    3:0.06
    4:0.06
    5:0.03
    6:0.09
    7:0.58
    Negative Logits
    <eos>
    -0.75
     (
    -0.74
    ↵↵
    -0.74
     
    -0.67
     I
    -0.66
    ,
    -0.66
     -
    -0.63
    .
    -0.63
     [
    -0.63
     and
    -0.63
    POSITIVE LOGITS
     itſelf
    1.29
     Theſe
    1.27
     Monfieur
    1.26
     myſelf
    1.22
     Efq
    1.20
     ―――――
    1.15
     Anſ
    1.14
     Majefty
    1.14
     doubtnut
    1.13
    ſelf
    1.11
    Activations Density 0.061%

    No Known Activations