© 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. 85091
    Prev
    Next
    INDEX
    Explanations

    commands or actions related to user interface navigation

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

    confirming or selecting actions

    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
     FormData
    -0.55
     queſta
    -0.54
     Мексичка
    -0.52
    principalTable
    -0.50
    labelledby
    -0.49
    Facades
    -0.49
     QtCore
    -0.48
     columnIndex
    -0.48
    IsMutable
    -0.47
    Хьажоргаш
    -0.47
    POSITIVE LOGITS
     Agreed
    0.34
     agreed
    0.33
     agree
    0.32
     Offisielt
    0.32
    agreed
    0.31
     Agree
    0.30
     UPON
    0.30
     agreeing
    0.30
     Ähn
    0.29
     teilnehmen
    0.29
    Activations Density 0.006%

    No Known Activations