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. 469
    Prev
    Next
    INDEX
    Explanations

    attends to metadata or formatting information represented by empty tokens from subsequent tokens

    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
    0:0.05
    1:0.10
    2:0.14
    3:0.16
    4:0.08
    5:0.09
    6:0.18
    7:0.17
    Negative Logits
     myſelf
    -5.78
     itſelf
    -5.59
     Efq
    -5.59
     Monfieur
    -5.22
     Theſe
    -5.16
     ―――――
    -5.09
     Jefus
    -4.97
     Majefty
    -4.84
    ^(@)
    -4.84
     iſt
    -4.84
    POSITIVE LOGITS
     
    3.11
    ,
    2.97
    .
    2.95
     (
    2.88
    ↵↵
    2.75
    -
    2.53
    ↵
    2.48
    :
    2.38
     "
    2.33
    '
    2.33
    Activations Density 0.098%

    No Known Activations