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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 38-GEMMASCOPE-RES-16K
    11. 11991
    Prev
    Next
    INDEX
    Explanations

    phrases that express pleas or requests

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unieke
    -0.54
    Referanser
    -0.52
     keenly
    -0.50
     nieco
    -0.49
    得多
    -0.48
     często
    -0.47
    各类
    -0.46
    様々な
    -0.45
     všetko
    -0.45
    częściej
    -0.45
    POSITIVE LOGITS
     guy
    0.66
     damn
    0.65
     wrong
    0.63
     stupid
    0.63
     goddamn
    0.59
     grandkids
    0.56
     darn
    0.56
     whole
    0.56
     others
    0.56
     kids
    0.54
    Activations Density 0.620%

    No Known Activations