© 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. 28517
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and notation related to formal definitions and specifications

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

    conspiracy charges

    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
     snippetHide
    -0.60
     disponibilités
    -0.52
     CURIAM
    -0.50
    ardless
    -0.50
     TextAppearance
    -0.50
    XmlAccessType
    -0.49
    xrTableCell
    -0.48
    CPtr
    -0.47
    awtextra
    -0.46
    findpost
    -0.45
    POSITIVE LOGITS
    ViewFeatures
    0.47
    تقاوى
    0.46
     '\\;'
    0.42
    msgSender
    0.41
     OFDb
    0.40
    AutoField
    0.39
    ControllerAdvice
    0.39
    endregion
    0.36
    ConverterFactory
    0.35
     createSlice
    0.35
    Activations Density 0.405%

    No Known Activations