Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 35561
    Prev
    Next
    INDEX
    Explanations

    phrases related to welcoming and greeting in social and business contexts

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

    welcoming and receiving

    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
     cô
    -0.34
    исленность
    -0.33
     fleuve
    -0.31
    quième
    -0.31
    Ground
    -0.30
     interfaz
    -0.30
     enriquec
    -0.29
    Spaces
    -0.29
    Structure
    -0.29
     interface
    -0.29
    POSITIVE LOGITS
     welcoming
    0.70
     welcomed
    0.60
    IVEREF
    0.60
     menyambut
    0.60
     للاسماء
    0.58
     welcome
    0.55
    webElementXpaths
    0.54
    迎
    0.54
    迎え
    0.53
    tanleria
    0.52
    Activations Density 0.042%

    No Known Activations