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. 3094
    Prev
    Next
    INDEX
    Explanations

    code structure and documentation elements, particularly in programming contexts

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

    Code constructors or initializers

    np_acts-logits-general · gemini-2.0-flash

    constructors and initializers

    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
    ChildScrollView
    -0.63
    tencent
    -0.50
    లాలు
    -0.48
     camas
    -0.48
    ✥
    -0.45
    宙
    -0.45
    atshirt
    -0.45
     digital
    -0.44
     running
    -0.42
     bed
    -0.41
    POSITIVE LOGITS
     constructor
    1.55
    constructor
    1.42
     constructors
    1.39
     Constructor
    1.38
     Constructors
    1.24
     constructeur
    1.18
    Constructor
    1.16
     构造
    1.09
     Constructs
    1.08
    构造
    1.08
    Activations Density 0.151%

    No Known Activations