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. 608
    Prev
    Next
    INDEX
    Explanations

    coding-related constructs and data structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    )++;
    -0.74
    ]--;
    -0.73
    ()]
    
    -0.66
     autorytatywna
    -0.64
    *)&
    -0.63
     *)&
    -0.63
    []):
    -0.63
    "){
    
    -0.62
     AssemblyTitle
    -0.60
    _();
    -0.59
    POSITIVE LOGITS
     purpoſe
    0.70
     preſent
    0.69
     leaſt
    0.68
    UnifiedTopology
    0.67
     ſtate
    0.67
     <<<<<<<<<<<<<<
    0.65
     ſtre
    0.64
     ſen
    0.63
     ſta
    0.63
     cauſe
    0.63
    Activations Density 0.409%

    No Known Activations