© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 486
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly semicolons and braces

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

    semicolons in code snippets

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

    statement terminators

    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
    Not in Any Lists

    No Comments

    Negative Logits
    TagHelper
    -0.80
    ssz
    -0.70
     تضيفلها
    -0.66
     utafitiHapana
    -0.65
     незавершена
    -0.65
     PASS
    -0.62
     oprot
    -0.60
     AppCompatTheme
    -0.59
     समीक्षाओं
    -0.58
     Sams
    -0.57
    POSITIVE LOGITS
    出版年
    0.74
    __;
    0.65
     mulighed
    0.65
    PerformLayout
    0.62
    endorong
    0.62
    SharedCtor
    0.61
     canadienne
    0.60
     noires
    0.60
     lumineuse
    0.59
     weiterer
    0.58
    Activations Density 0.038%

    No Known Activations