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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 29388
    Prev
    Next
    INDEX
    Explanations

    variations of the word "may" and its usage in sentences

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

    may and may not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     recensement
    -0.48
    <?
    -0.47
    setImageResource
    -0.46
    onViewCreated
    -0.45
    MLLoader
    -0.45
     IconData
    -0.45
     bottlenecks
    -0.44
    awtextra
    -0.44
    帖最后由
    -0.44
     disambiguazione
    -0.44
    POSITIVE LOGITS
    may
    0.66
     may
    0.66
     May
    0.56
     MAY
    0.56
    MAY
    0.55
     lawfully
    0.54
     shall
    0.52
    May
    0.52
    どうぞ
    0.52
     darf
    0.51
    Activations Density 0.050%

    No Known Activations