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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 114668
    Prev
    Next
    INDEX
    Explanations

    affirmative statements and expressions of quality or sincerity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     quelconque
    -0.41
    الحياه
    -0.39
    Boring
    -0.37
    Попис
    -0.37
    limsy
    -0.37
     Họ
    -0.37
    CodeAnalysis
    -0.37
    ilitary
    -0.36
     հղումներ
    -0.36
    ScopeManager
    -0.35
    POSITIVE LOGITS
     truly
    0.95
    truly
    0.82
    Truly
    0.79
     Truly
    0.78
     verdaderamente
    0.67
     verkligen
    0.66
    まさに
    0.66
     true
    0.65
    ChildScrollView
    0.64
    可谓
    0.62
    Activations Density 0.015%

    No Known Activations