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. 192
    Prev
    Next
    INDEX
    Explanations

    attends to tokens ending in "ny" from tokens that contain "an" or "ni."

    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
    0:0.61
    1:0.01
    2:0.03
    3:0.02
    4:0.14
    5:0.05
    6:0.02
    7:0.08
    Negative Logits
     un
    -0.35
     i
    -0.31
     s
    -0.29
    -
    -0.28
     le
    -0.27
     form
    -0.27
     cor
    -0.26
     origin
    -0.26
     il
    -0.26
     di
    -0.25
    POSITIVE LOGITS
     vectorielles
    0.53
    存于互联网档案馆
    0.52
    ImageContext
    0.50
     avoient
    0.50
     complètes
    0.49
     étrangères
    0.49
     particulières
    0.48
    usercontent
    0.47
    styleType
    0.47
     tartalomajánló
    0.47
    Activations Density 0.003%

    No Known Activations