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. 6042
    Prev
    Next
    INDEX
    Explanations

    attends to author initials from their corresponding periods or punctuation marks

    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
    0:0.34
    1:0.20
    2:0.12
    3:0.06
    4:0.07
    5:0.05
    6:0.06
    7:0.06
    Negative Logits
     dotenv
    -0.34
    𝗸
    -0.32
     kohdetta
    -0.31
    𝘃
    -0.29
     Ronnie
    -0.29
     '.',
    -0.29
    願い
    -0.29
     Fawcett
    -0.28
     Hewitt
    -0.28
    fontein
    -0.28
    POSITIVE LOGITS
    ReusableCell
    0.42
     gynhyrchwyd
    0.39
     خارجية
    0.38
    imetric
    0.33
     autorytatywna
    0.32
     Wiktionnaire
    0.32
    PMailer
    0.31
    TALL
    0.31
    Manbalar
    0.31
    coroutines
    0.31
    Activations Density 0.191%

    No Known Activations