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. 1177
    Prev
    Next
    INDEX
    Explanations

    attends to the token "<?" from arbitrary subsequent tokens

    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
    0:0.02
    1:0.69
    2:0.00
    3:0.01
    4:0.02
    5:0.01
    6:0.02
    7:0.04
    8:0.02
    9:0.02
    10:0.01
    11:0.01
    12:0.01
    13:0.01
    14:0.01
    15:0.01
    Negative Logits
     Económica
    -0.13
    Hauptartikel
    -0.11
    </strong>
    -0.11
    SPATH
    -0.10
     birlikte
    -0.10
     kultury
    -0.10
     środków
    -0.10
    jména
    -0.10
     zdarma
    -0.10
     serpiente
    -0.09
    POSITIVE LOGITS
    
    0.39
    SharedCtor
    0.39
    
    0.39
    
    0.38
    RegressionTest
    0.37
     betweenstory
    0.37
    
    0.36
    octa
    0.36
    Ay
    0.35
     tiguan
    0.35
    Activations Density 0.220%

    No Known Activations