© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 37961
    Prev
    Next
    INDEX
    Explanations

    references to political systems or the concept of a republic

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

    republic of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    visející
    -0.48
     estekak
    -0.41
    WINNER
    -0.41
    AddTagHelper
    -0.39
    Nate
    -0.38
     怎样
    -0.38
     poète
    -0.38
    balleur
    -0.37
     CreateTagHelper
    -0.37
     poème
    -0.37
    POSITIVE LOGITS
     republic
    0.88
     Republic
    0.87
    Republic
    0.79
     REPUBLIC
    0.65
     Republics
    0.63
     republics
    0.60
     repub
    0.60
    republic
    0.59
    jsxFileName
    0.55
    もしれない
    0.49
    Activations Density 0.003%

    No Known Activations