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

    Neuronpedia

    APICircuit 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. 20-GEMMASCOPE-RES-16K
    6. 2925
    Prev
    Next
    INDEX
    Explanations

    functional elements of a programming API related to resource management

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

    "Microsoft" or Microsoft products

    np_acts-logits-general · gemini-2.0-flash

    Microsoft or Azure API calls

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    __':
    -0.63
    __':
    
    -0.62
     NSCoder
    -0.61
    UserScript
    -0.57
     Dí
    -0.56
     NSObject
    -0.56
     ModelExpression
    -0.55
    Rabbi
    -0.55
     OkHttpClient
    -0.54
     bombers
    -0.52
    POSITIVE LOGITS
     Microsoft
    1.33
    Microsoft
    1.20
     Azure
    1.13
    microsoft
    1.10
     microsoft
    1.05
    Azure
    0.97
    azure
    0.95
     MS
    0.92
    微软
    0.92
     Xbox
    0.90
    Activations Density 0.483%

    No Known Activations