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. 1229
    Prev
    Next
    INDEX
    Explanations

    attends to the token "from" related to various subsequent framework 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
    0:0.02
    1:0.02
    2:0.03
    3:0.03
    4:0.04
    5:0.02
    6:0.03
    7:0.03
    8:0.02
    9:0.02
    10:0.02
    11:0.03
    12:0.52
    13:0.04
    14:0.03
    15:0.02
    Negative Logits
    <unused23>
    -2.34
    <unused17>
    -2.34
    <unused14>
    -2.34
    <unused16>
    -2.34
    [@BOS@]
    -2.34
    <pad>
    -2.34
    <unused3>
    -2.34
    <unused8>
    -2.34
    <unused28>
    -2.34
    <unused41>
    -2.34
    POSITIVE LOGITS
    ,
    0.71
     
    0.56
    -
    0.49
    '
    0.49
    .
    0.48
     (
    0.44
    ↵
    0.44
    :
    0.42
    ↵↵
    0.41
    )
    0.39
    Activations Density 0.015%

    No Known Activations