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. Residual Stream - 16k
    5. 16-GEMMASCOPE-RES-16K
    6. 9102
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to attention and its implications in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    "Att" or "ATT" acronym/token

    np_acts-logits-general · gemini-2.0-flash

    attending, attributing, attorney

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MethodManager
    -0.84
    //
    -0.82
     propOrder
    -0.81
    GEBURTSDATUM
    -0.77
    ScopeManager
    -0.75
     وتسجيلات
    -0.75
     jLabel
    -0.74
    '])){
    
    -0.72
    sizeCache
    -0.72
    ModelSerializer
    -0.70
    POSITIVE LOGITS
     Att
    0.87
     AT
    0.82
     att
    0.79
    att
    0.77
    Att
    0.76
    ATT
    0.75
     ATT
    0.74
    ATN
    0.73
     At
    0.72
     attr
    0.71
    Activations Density 0.494%

    No Known Activations