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. 0-GEMMASCOPE-RES-16K
    6. 6891
    Prev
    Next
    INDEX
    Explanations

    references to "ground" or "ground state" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.hook_resid_post
    Hook Layer
    0
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Groetjes
    -0.70
     Мексичка
    -0.67
     beginnetje
    -0.67
    Bises
    -0.64
    AndEndTag
    -0.64
    OGND
    -0.62
     ainfi
    -0.62
     remporté
    -0.62
    routeProvider
    -0.62
     vPvB
    -0.60
    POSITIVE LOGITS
     ground
    1.34
     Ground
    1.16
    Ground
    1.03
    ground
    1.00
     GROUND
    0.94
     security
    0.81
    security
    0.80
    GROUND
    0.77
     safety
    0.66
    gu
    0.61
    Activations Density 0.146%

    No Known Activations