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. 10751
    Prev
    Next
    INDEX
    Explanations

    phrases related to user capabilities and permissions in software

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

    ease or ability

    np_acts-logits-general · gemini-2.0-flash

    can/could do easily

    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
    waża
    -0.48
     Inscrivez
    -0.48
     endeavouring
    -0.45
     tratando
    -0.41
     autorisation
    -0.41
     mogen
    -0.41
     stenosis
    -0.41
     etmeye
    -0.41
    Har
    -0.41
     wajib
    -0.40
    POSITIVE LOGITS
     easily
    1.69
    easily
    1.49
     Easily
    1.34
    Easily
    1.31
     facilement
    1.28
     fácilmente
    1.17
     facilmente
    1.16
     freely
    1.14
     readily
    1.08
     comfortably
    1.05
    Activations Density 0.450%

    No Known Activations