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-2-9B-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 16098
    Prev
    Next
    INDEX
    Explanations

    references to gaming and computer hardware compatibility

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

    "board" in code or game contexts

    np_acts-logits-general · gemini-2.0-flash

    board data structures and games

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    createStatement
    -0.42
    ActivityResult
    -0.39
    Carriera
    -0.38
    PullParser
    -0.38
     MainAxisSize
    -0.38
     endforeach
    -0.38
     Principe
    -0.37
     thả
    -0.36
    êmio
    -0.35
     iload
    -0.35
    POSITIVE LOGITS
     board
    2.34
     Board
    2.08
     boards
    2.05
    Board
    2.02
    board
    1.98
     BOARD
    1.94
    BOARD
    1.81
     Boards
    1.80
    boards
    1.77
    Boards
    1.60
    Activations Density 0.124%

    No Known Activations