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. 11101
    Prev
    Next
    INDEX
    Explanations

    code snippets related to data return and processing logic in programming

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

    C code bit shifting and masking macros

    np_acts-logits-general · gemini-2.0-flash

    macros and bit shifts

    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
    __":
    
    -0.56
    }]{
    -0.56
    __':
    
    -0.55
    __":
    -0.54
     rospy
    -0.48
    AsUp
    -0.48
    ReusableCell
    -0.47
     kasarigan
    -0.46
    randomUUID
    -0.46
    __':
    -0.44
    POSITIVE LOGITS
     ((
    2.00
    ((
    1.94
     [(
    1.64
    [(
    1.59
    >((
    1.55
    (((
    1.50
    )((
    1.39
     (((
    1.35
    =((
    1.25
     (($
    1.21
    Activations Density 0.550%

    No Known Activations