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. 74
    Prev
    Next
    INDEX
    Explanations

    attends to the second token in a sequence from the first token in the sequence

    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
    0:0.03
    1:0.08
    2:0.49
    3:0.08
    4:0.07
    5:0.05
    6:0.06
    7:0.10
    Negative Logits
    .
    -1.19
    ,
    -1.16
     (
    -0.94
     in
    -0.93
    (
    -0.91
    -
    -0.91
    in
    -0.89
    1
    -0.87
    :
    -0.84
    ;
    -0.82
    POSITIVE LOGITS
     myſelf
    1.63
     itſelf
    1.61
     ―――――
    1.61
     Efq
    1.60
    ^(@)
    1.60
    NUMX
    1.56
     doubtnut
    1.54
     Forumite
    1.49
     ་་
    1.49
    >\<^
    1.48
    Activations Density 0.011%

    No Known Activations