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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 24-GEMMASCOPE-RES-16K
    6. 2292
    Prev
    Next
    INDEX
    Explanations

    elements of API function definitions and their structures

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

    Followed by an action verb

    np_acts-logits-general · gemini-2.0-flash

    performing code actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Removes
    -0.61
     removed
    -0.59
     removal
    -0.59
     Removed
    -0.58
     REMOV
    -0.57
     Removing
    -0.57
    Removing
    -0.55
    Removal
    -0.55
     sorting
    -0.55
    removing
    -0.55
    POSITIVE LOGITS
     execute
    1.00
     perform
    0.95
    execute
    0.92
     do
    0.88
     run
    0.85
    perform
    0.83
    do
    0.80
     Perform
    0.79
    exec
    0.79
    Execute
    0.78
    Activations Density 0.222%

    No Known Activations