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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 6419
    Prev
    Next
    INDEX
    Explanations

    attends to closing curly braces from arbitrary tokens

    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
    1:0.13
    2:0.24
    3:0.08
    4:0.10
    5:0.06
    6:0.07
    7:0.16
    Negative Logits
     AssemblyTitle
    -0.48
    erequisites
    -0.42
    ]='\
    -0.41
    ernalia
    -0.40
    erequisite
    -0.40
    ubehör
    -0.40
     Bader
    -0.39
    gebob
    -0.38
     سكانية
    -0.38
    loroethene
    -0.38
    POSITIVE LOGITS
    formik
    0.40
    azgo
    0.35
     dure
    0.34
    حوالہ
    0.32
    FormBuilder
    0.32
    Unity
    0.30
     vay
    0.30
     yea
    0.30
    multer
    0.30
     SwiftUI
    0.29
    Activations Density 0.024%

    No Known Activations