© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 3760
    Prev
    Next
    INDEX
    Explanations

    numerical values and color specifications in a CSS-like format

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    . by reducing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    Negative Logits
    (!__
    -0.52
     présence
    -0.47
     vaik
    -0.42
     aanbod
    -0.42
    ATTACH
    -0.41
     afhankelijk
    -0.41
     vícti
    -0.40
     financière
    -0.40
     rentrée
    -0.40
    Jereo
    -0.40
    POSITIVE LOGITS
    GOTREF
    0.79
    )':
    0.66
    httphttps
    0.64
     nawr
    0.63
    TagHelper
    0.62
     урна
    0.62
    manteau
    0.61
    elemField
    0.60
     autorytatywna
    0.60
    ̈́
    0.60
    Activations Density 0.031%

    No Known Activations