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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 4674
    Prev
    Next
    INDEX
    Explanations

    terms related to payment or financial transactions

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

    "pay" and related payment concepts

    np_acts-logits-general · gemini-2.0-flash

    paying for things or attention

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     autorytatywna
    -0.98
    __(/*!
    -0.94
    GEBURTSDATUM
    -0.93
     resourceCulture
    -0.87
     CreateTagHelper
    -0.83
    ✨:
    -0.81
     ModelExpression
    -0.78
     للاسماء
    -0.78
    EDEFAULT
    -0.78
     HasFactory
    -0.74
    POSITIVE LOGITS
     homage
    0.80
     attention
    0.74
     dividends
    0.72
     PAY
    0.71
     bills
    0.71
     tribute
    0.67
    Paying
    0.67
     pay
    0.64
    offs
    0.64
     Paying
    0.64
    Activations Density 0.071%

    No Known Activations