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. 6143
    Prev
    Next
    INDEX
    Explanations

    attends to the first token in the sequence from various coding-related 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
    0:0.09
    1:0.11
    2:0.10
    3:0.11
    4:0.08
    5:0.03
    6:0.26
    7:0.18
    Negative Logits
    GenerationType
    -0.36
     initComponents
    -0.36
    HostException
    -0.34
     createSprite
    -0.32
    ftagPool
    -0.30
    XtraEditors
    -0.30
    nastics
    -0.28
     Autorisations
    -0.28
    ElementException
    -0.28
    böz
    -0.27
    POSITIVE LOGITS
     Roskov
    0.32
     Opium
    0.31
    Autoritní
    0.31
     AssemblyCulture
    0.31
    بوابة
    0.31
    textAppearance
    0.31
     SEDS
    0.30
    Sinh
    0.30
    <bos>
    0.30
     Gaby
    0.29
    Activations Density 0.107%

    No Known Activations