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. 6045
    Prev
    Next
    INDEX
    Explanations

    attends to leadership-related tokens from corresponding style-related 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
    0:0.55
    1:0.01
    2:0.02
    3:0.03
    4:0.20
    5:0.10
    6:0.01
    7:0.03
    Negative Logits
     precisione
    -0.40
     sebou
    -0.40
     enfans
    -0.39
     originais
    -0.37
     imidlertid
    -0.36
     chrétien
    -0.35
     varandra
    -0.35
     sensazione
    -0.33
     militaires
    -0.33
     nødven
    -0.33
    POSITIVE LOGITS
     @"/
    0.42
    bukkit
    0.41
     autorytatywna
    0.39
    enderror
    0.37
     AssemblyTitle
    0.36
    omitempty
    0.35
    :✨
    0.35
     par
    0.34
    hithe
    0.34
    refront
    0.34
    Activations Density 0.008%

    No Known Activations