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. 5505
    Prev
    Next
    INDEX
    Explanations

    attends to numeric tokens followed by punctuation and other specific characters from subsequent tokens containing digits or punctuation

    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
    1:0.07
    2:0.06
    3:0.09
    4:0.48
    5:0.08
    6:0.05
    7:0.06
    Negative Logits
    AndEndTag
    -0.29
    payed
    -0.28
     fora
    -0.28
     worst
    -0.27
    ―
    -0.26
    jda
    -0.26
     transfieras
    -0.26
     članak
    -0.26
     Bib
    -0.26
     Byers
    -0.26
    POSITIVE LOGITS
    OGND
    0.39
    /**
    0.38
    ConstraintMaker
    0.37
    SequentialGroup
    0.35
    etcode
    0.34
     createState
    0.34
    AndroidJUnit
    0.33
    ethene
    0.33
    WriteTagHelper
    0.33
    Odkazy
    0.33
    Activations Density 1.759%

    No Known Activations