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. 1292
    Prev
    Next
    INDEX
    Explanations

    attends to the asterisk-marked tokens from the later tokens with brackets

    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
    0:0.03
    1:0.02
    2:0.00
    3:0.01
    4:0.01
    5:0.04
    6:0.30
    7:0.19
    8:0.02
    9:0.04
    10:0.02
    11:0.03
    12:0.03
    13:0.05
    14:0.06
    15:0.08
    Negative Logits
    <eos>
    -0.54
     Италијани
    -0.48
    |
    -0.47
     ویکی‌پدی
    -0.46
    \/
    -0.43
    KommentareTeilen
    -0.43
     |
    -0.42
    \"
    -0.42
    \
    -0.42
     resourceCulture
    -0.41
    POSITIVE LOGITS
     ferons
    0.29
    wahati
    0.27
     traseiro
    0.27
     enfans
    0.26
     收納
    0.26
     whoſe
    0.26
     abstrato
    0.25
    ⟬
    0.25
     plufieurs
    0.25
     Inscrivez
    0.24
    Activations Density 0.298%

    No Known Activations