© 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. 491
    Prev
    Next
    INDEX
    Explanations

    descriptions of food items, particularly highlighting textures and flavor profiles

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

    references to specific food items and their qualities in a menu context

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

    food and dishes

    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
    rato
    -0.47
    hya
    -0.44
    洋
    -0.43
    وله
    -0.41
    em
    -0.41
    ysql
    -0.39
     kaffe
    -0.38
    hms
    -0.38
    例文帳に追加
    -0.38
     veux
    -0.38
    POSITIVE LOGITS
    
    0.85
    DoubleQuotes
    0.84
    цездатний
    0.83
     pinulongan
    0.78
    تقاوى
    0.75
     HttpNotFound
    0.75
    ReusableCell
    0.73
     lenker
    0.71
    juana
    0.70
    rrggbb
    0.70
    Activations Density 0.082%

    No Known Activations