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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 105624
    Prev
    Next
    INDEX
    Explanations

    URL-like or path-like structures in text

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

    URL paths

    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
     Ruhe
    -0.36
    essentiel
    -0.34
     Gry
    -0.33
     TYP
    -0.32
     Explore
    -0.31
    tra
    -0.31
     idéia
    -0.30
    ví
    -0.30
    Minh
    -0.30
     verme
    -0.30
    POSITIVE LOGITS
    GEBURTSDATUM
    0.65
     للمعارف
    0.65
    RTSC
    0.64
    دانشنامهٔ
    0.62
    GIVEREF
    0.61
    
    0.61
    XtraGrid
    0.60
     CreateTagHelper
    0.58
     للاسماء
    0.58
    multer
    0.58
    Activations Density 0.001%

    No Known Activations