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. 6614
    Prev
    Next
    INDEX
    Explanations

    attends to ending punctuation tokens from tokens that introduce additional information or context

    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
    0:0.07
    1:0.09
    2:0.09
    3:0.44
    4:0.09
    5:0.04
    6:0.07
    7:0.07
    Negative Logits
     Waray
    -0.52
    Personendaten
    -0.47
    azgo
    -0.47
    IsContent
    -0.46
     समीक्षक
    -0.45
     lenker
    -0.44
     bigli
    -0.43
    principalColumn
    -0.42
     NDEBUG
    -0.42
    Datuak
    -0.41
    POSITIVE LOGITS
    udesta
    0.36
    +:+
    0.33
    AddTagHelper
    0.33
    程度上
    0.28
    \}\\
    0.28
    protoimpl
    0.28
    ynie
    0.27
    illies
    0.27
    /$
    0.26
    стях
    0.26
    Activations Density 1.252%

    No Known Activations