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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 113373
    Prev
    Next
    INDEX
    Explanations

    HTML and CSS code structures, particularly elements related to user interface layouts

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

    chat completions user

    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
    tagHelperRunner
    -0.58
     PhpStorm
    -0.47
    transQ
    -0.46
     Psicología
    -0.45
    SpringBootTest
    -0.45
    новништво
    -0.45
    abestanden
    -0.43
    nvarchar
    -0.43
     fels
    -0.42
    ⤒
    -0.42
    POSITIVE LOGITS
    jScrollPane
    0.86
    responsive
    0.85
     Responsive
    0.67
     responsive
    0.65
    Responsive
    0.60
    ImageContext
    0.54
    JScrollPane
    0.54
     JScrollPane
    0.52
     препратки
    0.50
    ChildScrollView
    0.50
    Activations Density 0.001%

    No Known Activations