© 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. 53418
    Prev
    Next
    INDEX
    Explanations

    guidance and tips for social interactions and etiquette

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

    giving tips and advice

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

    Phrases that introduce a list of suggestions, tips, rules, or instructional items (e.g., "here are," "the following," "tips," "rules").

    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
     Италијани
    -0.54
    ScopeManager
    -0.52
    pédie
    -0.51
     CascadeType
    -0.49
     noDo
    -0.47
     MessageBoxIcon
    -0.47
    SaveChangesAsync
    -0.47
     ModelExpression
    -0.47
    uestamente
    -0.46
     pintadas
    -0.46
    POSITIVE LOGITS
     tips
    0.62
     techniques
    0.60
     methods
    0.58
     recommendations
    0.56
     suggestions
    0.53
     tricks
    0.51
    的方法
    0.51
     steps
    0.50
     strategies
    0.50
     advice
    0.48
    Activations Density 0.015%

    No Known Activations