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. 10169
    Prev
    Next
    INDEX
    Explanations

    terms related to health funding and research support from organizations

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

    New Zealand locations and organizations

    np_acts-logits-general · gemini-2.0-flash

    New Zealand and Māori

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    ēju
    -0.58
    principalColumn
    -0.56
     thước
    -0.51
     peine
    -0.50
     Nigerian
    -0.49
    rhosis
    -0.49
    MENAFN
    -0.48
     ^
    -0.47
    țele
    -0.47
     Hershey
    -0.47
    POSITIVE LOGITS
     Maori
    1.08
     Auckland
    1.01
     Māori
    1.00
     NZ
    0.97
    NZ
    0.94
    Auckland
    0.94
     Christchurch
    0.91
     Dunedin
    0.85
    nz
    0.85
     Otago
    0.84
    Activations Density 0.100%

    No Known Activations