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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 162
    Prev
    Next
    INDEX
    Explanations

    phrases indicating successful performance or effectiveness

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

    Followed by adverbs of manner

    np_acts-logits-general · gemini-2.0-flash

    works well

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    ScopeManager
    -0.95
    BeginContext
    -0.85
    InvalidProtocol
    -0.83
     ComVisible
    -0.79
    TagMode
    -0.76
     Normdatei
    -0.76
    InjectAttribute
    -0.76
    RenderAtEndOf
    -0.75
    بوابة
    -0.73
    ValueStyle
    -0.70
    POSITIVE LOGITS
     fine
    0.82
     perfectly
    0.81
     best
    0.80
     well
    0.78
     better
    0.77
     wonders
    0.76
     merveille
    0.72
     bäst
    0.71
     perfect
    0.67
     flawlessly
    0.67
    Activations Density 0.128%

    No Known Activations