© 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. 20998
    Prev
    Next
    INDEX
    Explanations

    details about inscriptions and markings in documents or artworks

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

    marked and written information

    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
    ChildScrollView
    -0.33
     drivers
    -0.30
     controll
    -0.29
     bou
    -0.29
     philipp
    -0.28
     Isle
    -0.28
    PickerController
    -0.27
     divers
    -0.26
     rest
    -0.26
    よろしくお願いします
    -0.26
    POSITIVE LOGITS
     Normdatei
    0.71
     disambiguazione
    0.65
     AssemblyCompany
    0.65
    RegistryLite
    0.64
    лтамалар
    0.59
    tonode
    0.59
    ffilm
    0.59
     handwritten
    0.57
     ویکی‌پدی
    0.55
    MENAFN
    0.55
    Activations Density 0.009%

    No Known Activations