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

    Neuronpedia

    APICircuit 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. 30305
    Prev
    Next
    INDEX
    Explanations

    programming syntax and structure related to variable definitions and manipulations

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

    macros and programming definitions

    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
     
    -0.41
    ↵
    -0.38
     “
    -0.35
    ↵↵
    -0.33
     '
    -0.31
     recommandons
    -0.31
     recommandée
    -0.29
      
    -0.29
     Có
    -0.29
     "
    -0.28
    POSITIVE LOGITS
     autorytatywna
    1.00
     queſta
    0.90
     незавершена
    0.82
     transfieras
    0.81
    verwijspagina
    0.80
     esternos
    0.80
    
    0.77
    RTLR
    0.75
     nahilalakip
    0.75
     pinulongan
    0.72
    Activations Density 0.009%

    No Known Activations