Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 6801
    Prev
    Next
    INDEX
    Explanations

    attends to last names from corresponding preceding tokens that indicate family relationships or identities

    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
    0:0.47
    1:0.13
    2:0.10
    3:0.05
    4:0.06
    5:0.04
    6:0.04
    7:0.06
    Negative Logits
     فريبيس
    -0.32
     chi̍t
    -0.32
     Dwell
    -0.31
     iconFacebook
    -0.31
     للمعارف
    -0.31
    ftagPool
    -0.31
    vician
    -0.30
    oient
    -0.29
     doubtnut
    -0.29
     himſelf
    -0.28
    POSITIVE LOGITS
    verwijspagina
    0.37
    ScopeManager
    0.33
    ⓧ
    0.29
    +#+#
    0.24
    homonymie
    0.23
    OGND
    0.23
    ///</
    0.22
     MotionEvent
    0.22
    2
    0.22
     سكانية
    0.22
    Activations Density 0.066%

    No Known Activations