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. 6668
    Prev
    Next
    INDEX
    Explanations

    terms related to data structures and their manipulations in programming contexts

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

    "element" or "data" in code/tests

    np_acts-logits-general · gemini-2.0-flash

    element testing

    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
    CodeAttribute
    -0.51
     photo
    -0.48
    rift
    -0.46
     ozna
    -0.46
    kunt
    -0.43
     thân
    -0.43
    jema
    -0.42
    ായി
    -0.42
     Mitochondrial
    -0.42
    TypeMap
    -0.42
    POSITIVE LOGITS
     elements
    1.08
     element
    1.07
     elementos
    0.96
     éléments
    0.96
    elements
    0.94
     Element
    0.92
    element
    0.92
     elementi
    0.92
     Elements
    0.91
     elemento
    0.91
    Activations Density 0.502%

    No Known Activations