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. 10-GEMMASCOPE-ATT-16K
    6. 5803
    Prev
    Next
    INDEX
    Explanations

    attends to the token "href" from various tokens that include links

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.attn.hook_z
    Hook Layer
    10
    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
    0:0.07
    1:0.12
    2:0.10
    3:0.05
    4:0.05
    5:0.05
    6:0.06
    7:0.46
    Negative Logits
    findpost
    -0.43
     nakalista
    -0.43
     betweenstory
    -0.40
     مشين
    -0.40
    醐
    -0.34
    addGroup
    -0.33
    InSection
    -0.32
    ANTLR
    -0.32
    </thead>
    -0.32
     للاسماء
    -0.31
    POSITIVE LOGITS
     disambiguazione
    0.30
    ($("#
    0.27
     voulons
    0.26
     bordada
    0.25
    paddingVertical
    0.25
     affitto
    0.25
    "${
    0.24
     fysis
    0.24
    MemoryWarning
    0.24
     nordrhein
    0.23
    Activations Density 0.079%

    No Known Activations