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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 113342
    Prev
    Next
    INDEX
    Explanations

    punctuation and its context within dialogue or informal text

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

    speaking after punctuation

    np_acts-logits-general · gemini-2.5-flash-lite

    tokens that mark sentence boundaries or strong punctuation (sentence-ending periods, commas, quotes, exclamations and similar separators).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    ++
    -0.43
    ']}
    -0.43
     AssemblyCulture
    -0.42
    ']>
    -0.42
     stack
    -0.42
    >--}}
    -0.41
    \}}
    -0.41
    shund
    -0.40
    [:,:,
    -0.40
     Stack
    -0.40
    POSITIVE LOGITS
    principalColumn
    0.63
    addPreferredGap
    0.55
    AddTagHelper
    0.48
    HasForeignKey
    0.47
    featureID
    0.47
    Réponse
    0.45
    énario
    0.44
     Majefty
    0.44
     nahilalakip
    0.43
    setVerticalGroup
    0.42
    Activations Density 0.074%

    No Known Activations