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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 3213
    Prev
    Next
    INDEX
    Explanations

    function and method names related to programming and XML handling

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

    Code snippets or function-related terms

    np_acts-logits-general · gemini-2.0-flash

    programming function calls

    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
    хь
    -0.55
    ustu
    -0.53
    ={[
    -0.48
    hee
    -0.45
    ={{
    -0.44
     Art
    -0.42
    ValueStyle
    -0.41
    art
    -0.41
    ={
    
    -0.41
     gender
    -0.40
    POSITIVE LOGITS
    RectangleBorder
    0.85
     WaitForSeconds
    0.82
    StoryboardSegue
    0.78
     nahilalakip
    0.76
     Italijani
    0.74
     함수
    0.73
    AddTagHelper
    0.72
    
    0.70
    Искәрмәләр
    0.70
     informée
    0.69
    Activations Density 0.514%

    No Known Activations