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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 249
    Prev
    Next
    INDEX
    Explanations

    instances of formatting elements and structural components in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    (©
    -0.66
    LayoutPanel
    -0.66
    دانشنامهٔ
    -0.66
    _'.$
    -0.65
     chande
    -0.63
    AsUp
    -0.62
    expandindo
    -0.62
    RegressionTest
    -0.61
    뀝
    -0.60
     sánchez
    -0.57
    POSITIVE LOGITS
    httphttps
    0.59
     păr
    0.55
     calitate
    0.50
     indipendente
    0.47
     femininas
    0.45
     maș
    0.44
    NameInMap
    0.44
     afbeelding
    0.43
     verzame
    0.42
     çat
    0.42
    Activations Density 0.027%

    No Known Activations