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. 58796
    Prev
    Next
    INDEX
    Explanations

    terms related to spatial programming or UI framework components

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

    API endpoint names

    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
    tencent
    -0.39
     imp
    -0.38
    IComponent
    -0.37
     ST
    -0.36
    ↵↵
    -0.36
    ط
    -0.34
    as
    -0.33
     While
    -0.33
     as
    -0.33
      
    -0.32
    POSITIVE LOGITS
    OGND
    0.94
    parsedMessage
    0.92
    httphttps
    0.86
    GEBURTSDATUM
    0.85
     فريبيس
    0.82
    ंदीखरीदारी
    0.80
     صوتيه
    0.79
     Biôgrafia
    0.79
    expandindo
    0.78
     '\\;'
    0.77
    Activations Density 0.002%

    No Known Activations