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. 52999
    Prev
    Next
    INDEX
    Explanations

    terms associated with physical phenomena and scientific processes

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

    contribution is essential

    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
     Feldman
    -0.47
    érat
    -0.43
    cibility
    -0.43
    {~
    -0.42
     Marsh
    -0.42
    ="\
    -0.42
    idation
    -0.41
    Кре
    -0.40
    istik
    -0.40
    severance
    -0.40
    POSITIVE LOGITS
     contribution
    0.97
    GEBURTSDATUM
    0.97
     predominate
    0.90
     contributions
    0.90
     Contributions
    0.89
    Contribution
    0.87
     propOrder
    0.87
     contributo
    0.86
     contributing
    0.85
     Contribution
    0.85
    Activations Density 0.932%

    No Known Activations