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. 2899
    Prev
    Next
    INDEX
    Explanations

    specific instructions or steps related to a process

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

    Enumerations, bullets, and parenthesis

    np_acts-logits-general · gemini-2.0-flash

    symbols followed by words

    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
    ↵↵
    -0.60
    _
    -0.59
     vPvB
    -0.58
    .\\
    -0.56
    ,
    -0.56
    <eos>
    -0.55
    -
    -0.55
    /
    -0.54
    ↵
    -0.53
    ↵↵↵↵↵
    -0.53
    POSITIVE LOGITS
     The
    1.30
     There
    1.09
     Each
    0.96
    )";
    
    0.96
     When
    0.95
    interopRequire
    0.93
     They
    0.93
    "):
    
    0.93
     Those
    0.91
     Some
    0.91
    Activations Density 0.327%

    No Known Activations