Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 22-GEMMASCOPE-RES-16K
    6. 12959
    Prev
    Next
    INDEX
    Explanations

    references to "Pa" or "papaya" in technical or context-specific discussions

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

    "Pa" followed by other characters

    np_acts-logits-general · gemini-2.0-flash

    words starting with pa

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    IBOutlet
    -0.67
    WithEmail
    -0.59
    imų
    -0.56
     Wurzel
    -0.54
    ↩
    -0.54
     réflexion
    -0.53
     <<<<<<<<<<<<<<
    -0.52
    erialize
    -0.52
    ื่อง
    -0.52
    相关文章
    -0.51
    POSITIVE LOGITS
     PAH
    0.58
    parsedMessage
    0.58
    ########.
    0.57
    CodedInputStream
    0.56
     Paren
    0.55
     Opa
    0.53
    stateParams
    0.52
     jspb
    0.50
     Pa
    0.50
     PAX
    0.50
    Activations Density 0.098%

    No Known Activations