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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 46
    Prev
    Next
    INDEX
    Explanations

    underscores and specific formatting within code or structured text

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

    estatecat_piece, KFC_Forbidden, page_manager_contact_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
     étoient
    -0.39
     enfermo
    -0.36
     secundario
    -0.35
     reír
    -0.35
     mecánico
    -0.34
     étoit
    -0.34
     avoient
    -0.34
     telefónica
    -0.33
     malades
    -0.33
     Slovensku
    -0.32
    POSITIVE LOGITS
    twimg
    0.80
     المعيارى
    0.70
     فريبيس
    0.64
     createContext
    0.61
     queſto
    0.59
     للمعارف
    0.57
     sessionFactory
    0.57
     Transformations
    0.56
     незавершена
    0.56
    <unused40>
    0.56
    Activations Density 0.276%

    No Known Activations