© 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-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1574
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and related syntactic structures

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

    conditional statements and their related syntax in code

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

    the overshoot; enough to show; has launched

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     egentlig
    -0.48
     atque
    -0.44
    nloa
    -0.43
     børn
    -0.42
    setAccessible
    -0.42
    قایناقلار
    -0.41
    cedido
    -0.41
     refugi
    -0.41
     egentligen
    -0.41
    huit
    -0.40
    POSITIVE LOGITS
     err
    0.98
    DoubleQuotes
    0.82
    err
    0.79
     ok
    0.75
    AccessorTable
    0.71
    ArgumentParser
    0.69
     ModelExpression
    0.68
     OK
    0.68
    apimachinery
    0.68
    onActivityResult
    0.64
    Activations Density 0.018%

    No Known Activations