© 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-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 1801
    Prev
    Next
    INDEX
    Explanations

    references to database-related concepts or issues

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

    Second person perspective, advice, or suggestions

    np_acts-logits-general · gemini-2.0-flash

    technical terms or programming concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
    ↵
    -0.31
    ↵↵
    -0.29
     self
    -0.28
    PY
    -0.25
    ịnh
    -0.25
     inan
    -0.25
    1
    -0.23
     thousands
    -0.23
    ↵↵↵
    -0.23
     oldest
    -0.23
    POSITIVE LOGITS
    transQ
    0.96
    featureID
    0.91
     queſta
    0.87
     betweenstory
    0.82
    niſſe
    0.77
    ViewImports
    0.75
    MigrationBuilder
    0.74
    باردا
    0.70
    RTLR
    0.68
     nakalista
    0.68
    Activations Density 0.598%

    No Known Activations