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-2-9B
    4. Attention Out - 16k
    5. 0-GEMMASCOPE-ATT-16K
    6. 1176
    Prev
    Next
    INDEX
    Explanations

    attends to the token "everything" from various corresponding tokens that provide context or describe conditions

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-att/layer_0/width_16k/average_l0_61
    Prompts (Dashboard)
    16,384 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
    0:0.01
    1:0.01
    2:0.01
    3:0.02
    4:0.01
    5:0.01
    6:0.30
    7:0.18
    8:0.01
    9:0.01
    10:0.01
    11:0.04
    12:0.01
    13:0.17
    14:0.12
    15:0.01
    Negative Logits
     is
    -0.18
     Absicht
    -0.17
     Kräfte
    -0.16
     Zwecke
    -0.15
     from
    -0.14
     actualidad
    -0.14
     Hinweise
    -0.14
     Spitze
    -0.14
     totalidad
    -0.14
     eigentlichen
    -0.14
    POSITIVE LOGITS
    <unused43>
    0.95
    <unused80>
    0.94
    <pad>
    0.94
    <unused41>
    0.94
    <unused20>
    0.94
    <unused17>
    0.94
    <unused23>
    0.94
    <unused28>
    0.94
    <unused3>
    0.94
    <unused8>
    0.94
    Activations Density 0.049%

    No Known Activations