© 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 EvalsExportsAPI 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. 10665
    Prev
    Next
    INDEX
    Explanations

    references to threats and violent actions

    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
     hammering
    -0.54
     Paglinawan
    -0.51
     beating
    -0.49
    providedIn
    -0.47
     destrucción
    -0.45
     banging
    -0.45
     destruction
    -0.44
     Zerst
    -0.44
    WritableDatabase
    -0.43
     slamming
    -0.43
    POSITIVE LOGITS
     hurt
    0.72
     harm
    0.58
     harmed
    0.56
    hurt
    0.49
     seriously
    0.47
     Harm
    0.47
     ske
    0.45
     electro
    0.43
    harm
    0.43
    Harm
    0.43
    Activations Density 0.349%

    No Known Activations