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-2-2B
    4. Attention Out - 16k
    5. 10-GEMMASCOPE-ATT-16K
    6. 6012
    Prev
    Next
    INDEX
    Explanations

    attends to verbs indicating obtaining or acquiring from their respective locations

    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
    2:0.37
    3:0.07
    4:0.04
    5:0.04
    6:0.07
    7:0.10
    Negative Logits
     themselves
    -0.28
     muka
    -0.27
    enschappelijke
    -0.26
    im
    -0.26
    yot
    -0.25
    逅
    -0.25
    InView
    -0.25
     _{\
    -0.24
     tả
    -0.24
     ktorá
    -0.24
    POSITIVE LOGITS
     @"/
    0.37
    存于互联网档案馆
    0.37
     AssemblyVersion
    0.33
    CopyWith
    0.32
    TagMode
    0.31
     intents
    0.31
    icose
    0.30
    HomeAsUpEnabled
    0.30
    bootstrapcdn
    0.30
     Alegre
    0.29
    Activations Density 0.014%

    No Known Activations