© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 113029
    Prev
    Next
    INDEX
    Explanations

    references to lattice structures in mathematical or physical contexts

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

    lattice structure and components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    endforeach
    -0.49
    Karan
    -0.47
     Cesar
    -0.47
    
    -0.46
    Won
    -0.45
    出版年
    -0.45
    putstr
    -0.45
    Span
    -0.44
     المعيارى
    -0.44
     Karan
    -0.43
    POSITIVE LOGITS
     lattice
    0.98
     Lattice
    0.88
    Lattice
    0.88
    lattice
    0.73
     lattices
    0.66
     grid
    0.65
     решет
    0.59
     trellis
    0.59
     réseaux
    0.58
    راسیون
    0.56
    Activations Density 0.014%

    No Known Activations