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. 7522
    Prev
    Next
    INDEX
    Explanations

    references to the game "Candy Crush" and related gaming concepts

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

    references to popular mobile games and gaming terminology.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot

    Mobile game-related words

    np_acts-logits-general · gemini-2.0-flash

    mobile games

    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
     فريبيس
    -0.78
    ContentAsync
    -0.65
    AutoresizingMask
    -0.62
    WithIOException
    -0.61
    enderror
    -0.59
    rungsseite
    -0.58
    ymce
    -0.56
    --*/
    -0.53
    ologues
    -0.53
    SpringBootTest
    -0.53
    POSITIVE LOGITS
     Clash
    0.60
     tapping
    0.58
     IAP
    0.56
     Tap
    0.56
     Swipe
    0.55
     leaderboard
    0.54
     Leaderboard
    0.54
     levels
    0.54
    ALLENG
    0.54
     tap
    0.53
    Activations Density 0.151%

    No Known Activations