Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 10-GEMMASCOPE-ATT-16K
    6. 6599
    Prev
    Next
    INDEX
    Explanations

    attends to entities or concepts like questions or goals from tokens that contribute to defining those entities or concepts

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.attn.hook_z
    Hook Layer
    10
    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
    0:0.13
    1:0.15
    2:0.12
    3:0.15
    4:0.17
    5:0.03
    6:0.08
    7:0.13
    Negative Logits
    Significado
    -0.21
    gameserver
    -0.21
    --
    -0.20
    NegativeButton
    -0.20
     Signalez
    -0.20
    GeoNames
    -0.20
     including
    -0.19
    Parcelize
    -0.19
     ویکی‌پدی
    -0.19
    =!
    -0.18
    POSITIVE LOGITS
     nakalista
    0.38
     Wikispecies
    0.31
     kasarigan
    0.31
    '])->
    0.30
     endforeach
    0.28
    IVEREF
    0.28
    الحياه
    0.28
     ***!
    0.28
     autorytatywna
    0.27
    Története
    0.27
    Activations Density 0.578%

    No Known Activations