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. 4795
    Prev
    Next
    INDEX
    Explanations

    numeric statistics in a sports context

    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
     Several
    -0.52
    有多
    -0.50
    Several
    -0.49
    sient
    -0.48
    げん
    -0.47
     multiple
    -0.47
    tées
    -0.47
    asibility
    -0.47
    ssan
    -0.47
    opher
    -0.47
    POSITIVE LOGITS
    LookAnd
    0.86
     celor
    0.65
    XmlAccessorType
    0.64
    EndContext
    0.64
    tagHelperRunner
    0.63
    ParallelGroup
    0.63
    
    0.62
    forChild
    0.60
    RectangleBorder
    0.60
    sizeCache
    0.59
    Activations Density 0.270%

    No Known Activations