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-2-9B
    4. Attention Out - 16k
    5. 0-GEMMASCOPE-ATT-16K
    6. 1240
    Prev
    Next
    INDEX
    Explanations

    attends to the greeting "Dear" from various instances of addressed tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-att/layer_0/width_16k/average_l0_61
    Prompts (Dashboard)
    16,384 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
    0:0.01
    1:0.01
    2:0.08
    3:0.02
    4:0.00
    5:0.00
    6:0.08
    7:0.08
    8:0.01
    9:0.01
    10:0.01
    11:0.03
    12:0.05
    13:0.43
    14:0.07
    15:0.02
    Negative Logits
    SequentialGroup
    -0.70
     surla
    -0.62
    AndEndTag
    -0.62
    aarrggbb
    -0.60
     <<<<<<<<<<<<<<
    -0.60
    WithIOException
    -0.60
     ModelExpression
    -0.59
    webElementXpaths
    -0.58
    IUrlHelper
    -0.58
    rungsseite
    -0.57
    POSITIVE LOGITS
    0
    0.26
    7
    0.20
    3
    0.20
    s
    0.19
    d
    0.18
    6
    0.17
    4
    0.17
    m
    0.17
    Type
    0.17
    1
    0.17
    Activations Density 0.017%

    No Known Activations