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 - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 52958
    Prev
    Next
    INDEX
    Explanations

    high activation phrases or concepts related to conflict or tension, particularly during historical contexts

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

    Starts sentences with capital words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ervan
    -0.66
     dermed
    -0.65
     daarvan
    -0.64
    Enllaces
    -0.64
     оно
    -0.58
     daarmee
    -0.56
     cioè
    -0.56
     endforeach
    -0.56
    XDECREF
    -0.55
    OfWork
    -0.54
    POSITIVE LOGITS
    ंदीखरीदारी
    0.83
     kasarigan
    0.78
     cref
    0.73
     CreateTagHelper
    0.69
    WriteTagHelper
    0.65
    "}},
    0.65
    PerformLayout
    0.64
    "};
    0.64
    ,:);
    0.64
    StoreMessageInfo
    0.63
    Activations Density 0.274%

    No Known Activations