© 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
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 7119
    Prev
    Next
    INDEX
    Explanations

    symbols, formatting marks, and annotations often found in technical documents

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

    special characters used in scientific notation, references, and code symbols like '#', '^', '!', '\*', and '//'.

    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
    enumi
    -0.52
     
    -0.43
    Toole
    -0.40
    IContainer
    -0.39
    ParallelGroup
    -0.38
    mml
    -0.38
    {
    
    -0.38
     penutup
    -0.37
    section
    -0.37
    Brien
    -0.36
    POSITIVE LOGITS
     kasarigan
    0.73
    webtoken
    0.65
    OGND
    0.61
    jsxFileName
    0.60
     defaultstate
    0.60
     betweenstory
    0.57
    MessageTagHelper
    0.56
    
    0.56
    Derbyniad
    0.55
    elemField
    0.54
    Activations Density 1.266%

    No Known Activations