© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 10046
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and structural elements in equations or matrices

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

    mathematical matrices and tabular data formatting.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
    /**
    -0.46
    Географи
    -0.44
    /*++
    -0.40
     archaeologists
    -0.40
    clusal
    -0.39
    انتهای
    -0.39
    enderror
    -0.39
    IsMutable
    -0.38
     gills
    -0.38
    
    -0.38
    POSITIVE LOGITS
    ArrowToggle
    0.51
    setVerticalGroup
    0.49
    MessageOf
    0.45
     оригіналу
    0.43
    IsContent
    0.42
    GenerationType
    0.42
     Usaha
    0.39
     MainAxisSize
    0.39
    InitStruct
    0.38
     contextLoads
    0.37
    Activations Density 0.266%

    No Known Activations