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. 6164
    Prev
    Next
    INDEX
    Explanations

    attends to specific punctuation marks within a verbal phrase from linked conjunctions or elements earlier in the sequence

    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
    0:0.45
    1:0.24
    2:0.06
    3:0.05
    4:0.04
    5:0.01
    6:0.05
    7:0.06
    Negative Logits
     فريبيس
    -0.53
    มาะ
    -0.30
     صوتيه
    -0.29
     الرياضيه
    -0.29
    ConstraintMaker
    -0.29
    gnore
    -0.28
    TemporalType
    -0.28
    arture
    -0.28
    BagConstraints
    -0.27
     AssemblyCulture
    -0.27
    POSITIVE LOGITS
     lenker
    0.38
     Meksiku
    0.36
     VIDEOTAPE
    0.33
    JpaRepository
    0.31
     }{@
    0.31
    ThroughAttribute
    0.30
    Composable
    0.30
    raisemb
    0.29
    Népesség
    0.28
     disambiguazione
    0.28
    Activations Density 0.304%

    No Known Activations