Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 14612
    Prev
    Next
    INDEX
    Explanations

    elements and functions related to programming and data manipulation in various languages

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

    Code-related terms; datetime, vector, cli, Getter

    np_acts-logits-general · gemini-2.0-flash

    parsing date and time strings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tagHelper
    -0.48
    +'</
    -0.47
     ddelweddau
    -0.45
     Thru
    -0.44
    "]];
    -0.44
     mtrl
    -0.44
     ')[
    -0.43
    ='';
    
    -0.43
    tiefel
    -0.43
     Hälfte
    -0.42
    POSITIVE LOGITS
    Diwedd
    1.25
     AssemblyProduct
    1.13
    دانشنامهٔ
    1.09
    CloseOperation
    1.07
    Portály
    1.04
    UnsafeEnabled
    1.01
    GeoNames
    1.00
     ویکی‌پدیای
    0.97
    Datuak
    0.94
     Baillargeon
    0.92
    Activations Density 0.409%

    No Known Activations