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. 156
    Prev
    Next
    INDEX
    Explanations

    attends to the token "the" from various preceding tokens

    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
    0:0.04
    1:0.10
    2:0.06
    3:0.08
    4:0.12
    5:0.05
    6:0.12
    7:0.39
    Negative Logits
     myſelf
    -1.00
     greateſt
    -0.97
     Monfieur
    -0.95
     Efq
    -0.94
     itſelf
    -0.90
     faſt
    -0.90
     pleaſure
    -0.88
     purpoſe
    -0.88
     beſt
    -0.87
     poffible
    -0.85
    POSITIVE LOGITS
     e
    0.41
     di
    0.41
     v
    0.39
     or
    0.38
     me
    0.38
     por
    0.38
     con
    0.38
     c
    0.37
     que
    0.37
     can
    0.36
    Activations Density 2.656%

    No Known Activations