© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1814
    Prev
    Next
    INDEX
    Explanations

    technical terms and keywords related to data handling and processing

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

    configuration parameters

    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
     
    -0.58
     F
    -0.46
     De
    -0.45
    ,
    -0.45
     S
    -0.44
     T
    -0.43
     a
    -0.43
     Lo
    -0.43
    dom
    -0.43
     K
    -0.42
    POSITIVE LOGITS
    ConstraintMaker
    1.37
     فريبيس
    1.28
     myſelf
    1.27
     itſelf
    1.14
     uſed
    1.01
     pinulongan
    1.01
     deſt
    0.99
    +#+#
    0.97
     Jefus
    0.95
     raiſ
    0.95
    Activations Density 0.572%

    No Known Activations