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. 6068
    Prev
    Next
    INDEX
    Explanations

    This attention head attends to punctuation marks and grammatical connectors from subsequent content words.

    oai_attention-head · claude-3-7-sonnet-20250219Triggered by @trainjumperx

    attends to the token "refrigerator" from the token "safe."

    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
    1:0.14
    2:0.11
    3:0.10
    4:0.10
    5:0.03
    6:0.17
    7:0.22
    Negative Logits
    OfWeek
    -0.30
     Paglinawan
    -0.28
    djangoproject
    -0.28
    windowFixed
    -0.27
    ->_
    -0.27
     Greenfield
    -0.27
     createState
    -0.27
    __.__
    -0.26
    amek
    -0.26
    ViewFeatures
    -0.26
    POSITIVE LOGITS
    Xna
    0.27
    sizeCache
    0.25
     مشارکت‌کنندگان
    0.25
    ętr
    0.25
     mihi
    0.24
    nestjs
    0.24
     kuten
    0.23
     pinion
    0.23
     <=",
    0.23
    nocześnie
    0.22
    Activations Density 0.006%

    No Known Activations