© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 33-GEMMASCOPE-RES-16K
    6. 11855
    Prev
    Next
    INDEX
    Explanations

    references to curfews and time-related restrictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     maș
    -0.50
     puteți
    -0.50
     contactez
    -0.48
     exécut
    -0.48
     ejecut
    -0.47
     față
    -0.46
    țion
    -0.45
     behaviors
    -0.45
     Afterward
    -0.44
     uș
    -0.43
    POSITIVE LOGITS
    PTI
    0.68
     PTI
    0.66
     disambiguazione
    0.65
     UNI
    0.57
     kasarigan
    0.56
     Incidentally
    0.56
     `.
    0.55
     surla
    0.54
    帖最后由
    0.54
     humour
    0.54
    Activations Density 0.071%

    No Known Activations