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. 14524
    Prev
    Next
    INDEX
    Explanations

    terms related to athletic performance metrics and physical attributes

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

    Player skill or talent evaluation

    np_acts-logits-general · gemini-2.0-flash

    traits and abilities

    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
    BeginContext
    -0.65
    (!__
    -0.63
    GeneratedCode
    -0.59
    expandindo
    -0.54
     cœurs
    -0.53
    IntoConstraints
    -0.53
     automatiques
    -0.52
    tamos
    -0.52
    ubehör
    -0.52
    readyState
    -0.52
    POSITIVE LOGITS
     versatile
    0.67
     versatility
    0.67
     raw
    0.65
     versi
    0.60
    Versatile
    0.59
     explosive
    0.58
    vers
    0.57
     proven
    0.56
     ability
    0.56
     Poss
    0.56
    Activations Density 0.141%

    No Known Activations