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. 475
    Prev
    Next
    INDEX
    Explanations

    attends to the token "else" from tokens that follow it, reflecting conditional branching in programming structures

    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
    0:0.81
    1:0.01
    2:0.01
    3:0.03
    4:0.06
    5:0.01
    6:0.01
    7:0.03
    Negative Logits
    <eos>
    -0.39
     recently
    -0.32
     (
    -0.32
     if
    -0.27
     {
    -0.27
     of
    -0.27
     they
    -0.26
     चाहिए
    -0.26
     I
    -0.26
     since
    -0.25
    POSITIVE LOGITS
    SpringRunner
    0.49
    enumi
    0.48
    abestanden
    0.48
     Anſ
    0.46
     Wikimedijinoj
    0.45
    Viitteet
    0.44
     ―――――
    0.44
     متعلقه
    0.44
     uſe
    0.43
    neſs
    0.43
    Activations Density 0.011%

    No Known Activations