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. 0-GEMMASCOPE-ATT-16K
    6. 300
    Prev
    Next
    INDEX
    Explanations

    attends to opening square brackets from closing square brackets in mathematical expressions or code structures

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_0/width_16k/average_l0_104
    Prompts (Dashboard)
    36,864 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
    0:0.21
    1:0.02
    2:0.03
    3:0.03
    4:0.35
    5:0.14
    6:0.06
    7:0.12
    Negative Logits
    rungsseite
    -0.41
     متعلقه
    -0.39
    newUser
    -0.36
    `,`
    -0.35
    RegressionTest
    -0.34
     незавершена
    -0.34
    )$
    -0.33
    $
    -0.33
    begin
    -0.33
    covo
    -0.32
    POSITIVE LOGITS
    ConstraintMaker
    0.49
    UserScript
    0.47
    NSNotification
    0.44
    resizingMask
    0.43
    argout
    0.42
    +#+
    0.40
    glBindBuffer
    0.40
    newtheorem
    0.39
    migrationBuilder
    0.39
     WaitForSeconds
    0.39
    Activations Density 0.073%

    No Known Activations