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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 16354
    Prev
    Next
    INDEX
    Explanations

    patterns and specific numerical data formats

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    OGND
    -0.63
    httphttps
    -0.62
     Meksiku
    -0.59
     Савезне
    -0.55
    Tembelea
    -0.52
    AddTagHelper
    -0.51
     chi̍t
    -0.47
     мәкал
    -0.46
     ModelExpression
    -0.45
    RectangleBorder
    -0.45
    POSITIVE LOGITS
    HtmlAttribute
    0.45
     Mellon
    0.44
    ceres
    0.44
    setEditable
    0.42
    Jag
    0.42
    Characteristic
    0.41
    rible
    0.41
     IES
    0.41
    IE
    0.41
     SOS
    0.41
    Activations Density 0.007%

    No Known Activations