© 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 - 16k
    9. 22-GEMMASCOPE-RES-16K
    11. 7497
    Prev
    Next
    INDEX
    Explanations

    terms related to data types and binary structures, indicating an interest in technical or programming concepts

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

    Japanese or Japanese-like words

    np_acts-logits-general · gemini-2.0-flash

    words ending in ou

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Walton
    -0.77
    Хи
    -0.72
     Geno
    -0.70
    ReusableCell
    -0.70
     Chim
    -0.69
     Batavia
    -0.69
     Delf
    -0.68
    Gale
    -0.66
     Хи
    -0.64
     McCulloch
    -0.63
    POSITIVE LOGITS
     Kou
    1.21
     Cou
    1.14
     Dou
    1.13
     Brou
    1.08
     Chou
    1.07
    Kou
    1.06
     dou
    1.05
     Rou
    1.05
     COU
    1.05
     cou
    1.05
    Activations Density 0.610%

    No Known Activations