© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 113525
    Prev
    Next
    INDEX
    Explanations

    concepts related to safety and refuge in physical spaces

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

    sanctuary, refuge, oasis, paradise

    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
     المعيارى
    -0.64
     للمعارف
    -0.52
     jsPsych
    -0.49
     informée
    -0.48
    Бахар
    -0.48
    ndice
    -0.48
     Curse
    -0.47
     Administrativna
    -0.47
    Чыгана
    -0.46
    payee
    -0.46
    POSITIVE LOGITS
     sanctuary
    0.50
    #
    0.47
     spaces
    0.45
     refuge
    0.42
     moments
    0.39
     space
    0.39
     refugio
    0.38
    centralwidget
    0.37
    spaces
    0.36
     zone
    0.36
    Activations Density 0.009%

    No Known Activations