© 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. 312
    Prev
    Next
    INDEX
    Explanations

    references to collaborative structures and relationships within data

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

    terms related to keyboard layout and functionality

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

    lines, nodes, or copies

    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
    ÉR
    -0.48
     navires
    -0.45
     épaules
    -0.45
     débats
    -0.42
     vœux
    -0.42
     prochaines
    -0.42
    chre
    -0.42
     vaisseaux
    -0.42
    Sock
    -0.41
    ذك
    -0.40
    POSITIVE LOGITS
     betweenstory
    0.94
     виправивши
    0.89
    RenderAtEndOf
    0.88
    tagHelperRunner
    0.87
    OGND
    0.86
    ergies
    0.84
    出版年
    0.83
     referenties
    0.83
     Савезне
    0.82
     незавершена
    0.82
    Activations Density 0.949%

    No Known Activations