Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 24-GEMMASCOPE-RES-16K
    6. 8181
    Prev
    Next
    INDEX
    Explanations

    references to large numbers or quantifiable elements related to gaming experiences

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

    followed by aspects of gameplay or game design

    np_acts-logits-general · gemini-2.0-flash

    gameplay and game mechanics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ẵn
    -0.61
    ritation
    -0.60
    assertRaises
    -0.52
    isContained
    -0.51
    結論
    -0.51
     Meilleur
    -0.51
    رشف
    -0.51
     PhpStorm
    -0.50
     saluti
    -0.50
    ophanes
    -0.50
    POSITIVE LOGITS
     gameplay
    0.88
     traversal
    0.78
     player
    0.76
     environments
    0.73
     exploration
    0.71
     giocatore
    0.67
     playable
    0.66
     proced
    0.66
     mechanics
    0.65
     RPG
    0.65
    Activations Density 0.188%

    No Known Activations