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. 1170
    Prev
    Next
    INDEX
    Explanations

    attends to tokens that are related to the initialization of functions or methods from later 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
    0:0.01
    1:0.01
    2:0.57
    3:0.02
    4:0.00
    5:0.00
    6:0.03
    7:0.02
    8:0.02
    9:0.01
    10:0.01
    11:0.00
    12:0.01
    13:0.02
    14:0.17
    15:0.01
    Negative Logits
     {
    -0.44
    ↵
    -0.42
    this
    -0.41
     
    -0.40
    e
    -0.40
    var
    -0.39
    i
    -0.37
    mathrm
    -0.35
    R
    -0.35
     {
    
    -0.34
    POSITIVE LOGITS
     myſelf
    0.61
     purpoſe
    0.59
     ſind
    0.59
     ſont
    0.58
     faſt
    0.57
    majánló
    0.56
     ſta
    0.56
     autorytatywna
    0.54
     Houſe
    0.54
     ſte
    0.54
    Activations Density 0.018%

    No Known Activations