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. 1345
    Prev
    Next
    INDEX
    Explanations

    attends to tokens that include the pattern "Has" from subsequent tokens that include "min" and similar patterns

    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
    2:0.13
    3:0.45
    4:0.01
    5:0.01
    6:0.03
    7:0.03
    8:0.00
    9:0.01
    10:0.01
    11:0.01
    12:0.02
    13:0.07
    14:0.11
    15:0.01
    Negative Logits
    ign
    -0.77
    TR
    -0.75
    cul
    -0.74
    wor
    -0.74
    hy
    -0.74
    lor
    -0.74
    ses
    -0.73
    hip
    -0.73
    fer
    -0.73
    ser
    -0.73
    POSITIVE LOGITS
     civilización
    0.45
     plufieurs
    0.45
     Inscrivez
    0.44
     Económica
    0.44
     enfans
    0.43
     Italij
    0.42
     valentín
    0.41
     feroit
    0.40
     fotográfico
    0.39
     avoient
    0.39
    Activations Density 0.131%

    No Known Activations