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. 1932
    Prev
    Next
    INDEX
    Explanations

    phrases related to regulatory compliance and adherence to guidelines

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    PostExecute
    -0.66
    abestanden
    -0.61
    లాలు
    -0.59
     nakalista
    -0.58
    addPreferredGap
    -0.56
     насељу
    -0.55
    حياته
    -0.55
    adpleegd
    -0.54
    žku
    -0.53
    indd
    -0.50
    POSITIVE LOGITS
     rules
    1.08
     guidelines
    0.97
     requirements
    0.95
     norms
    0.92
     standards
    0.91
     regulations
    0.90
     principles
    0.86
     expectations
    0.85
     tradition
    0.85
     instructions
    0.84
    Activations Density 0.556%

    No Known Activations