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. 1196
    Prev
    Next
    INDEX
    Explanations

    attends to the token "year" related to the current or previous year from "year" tokens related to the subsequent year

    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
    1:0.02
    2:0.01
    3:0.07
    4:0.51
    5:0.02
    6:0.03
    7:0.03
    8:0.02
    9:0.01
    10:0.02
    11:0.02
    12:0.11
    13:0.02
    14:0.01
    15:0.01
    Negative Logits
     zijne
    -0.35
     year
    -0.34
     año
    -0.34
     året
    -0.31
    '):
    
    -0.31
     année
    -0.30
     nôtre
    -0.29
    "):
    
    -0.28
     anuales
    -0.28
    ()]
    
    -0.27
    POSITIVE LOGITS
     seg
    0.42
    metter
    0.41
     <
    0.39
     =
    0.39
    aser
    0.39
     TRANSMISSION
    0.39
     ADMINISTRATION
    0.38
     gol
    0.38
     ویکی‌پدی
    0.38
    ciel
    0.38
    Activations Density 0.051%

    No Known Activations