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-2-9B
    4. Residual Stream - 131k
    5. 32-GEMMASCOPE-RES-131K
    6. 129873
    Prev
    Next
    INDEX
    Explanations

    references to historical scholars and works

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

    historian name

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.32.hook_resid_post
    Hook Layer
    32
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Referanser
    -0.40
    
    -0.40
     吋
    -0.39
     operator
    -0.39
    娘娘
    -0.37
    Ellie
    -0.37
     finance
    -0.37
    Freddie
    -0.36
     кан
    -0.36
    sendTo
    -0.35
    POSITIVE LOGITS
     historian
    0.94
     historians
    0.86
     biographer
    0.81
     Historian
    0.75
     author
    0.70
     scholars
    0.69
     authors
    0.69
     scholar
    0.69
    scholar
    0.63
     sociologist
    0.63
    Activations Density 0.035%

    No Known Activations