© 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. 13-GEMMASCOPE-RES-16K
    11. 9492
    Prev
    Next
    INDEX
    Explanations

    references to violent actions, specifically attacks and their contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 13-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_13/width_16k/average_l0_132
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.13.hook_resid_post
    Hook Layer
    13
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     zelve
    -0.54
    rungsseite
    -0.52
     Picchu
    -0.49
     acreditar
    -0.45
     ſein
    -0.45
    sidemargin
    -0.45
     quiser
    -0.43
     uLocal
    -0.43
    躇
    -0.43
    ❉
    -0.41
    POSITIVE LOGITS
     attack
    1.63
    attack
    1.48
     Attack
    1.41
     attacks
    1.31
     ATTACK
    1.28
    Attack
    1.27
    ATTACK
    1.22
     attacking
    1.21
     attacked
    1.17
     ataque
    1.13
    Activations Density 0.200%

    No Known Activations