© 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-IT
    7. Residual Stream - 131k
    9. 9-GEMMASCOPE-RES-131K
    11. 35920
    Prev
    Next
    INDEX
    Explanations

    explicit and aggressive language directed at individuals

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

    "you" "idiot" "kill" "senseless"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AnimationsModule
    -0.64
     وتسجيلات
    -0.57
    WebElementEntity
    -0.51
    posedge
    -0.51
    XmlAccessType
    -0.50
    новниш
    -0.50
     referenties
    -0.48
     Vergrößern
    -0.48
    CommonModule
    -0.47
     Obrador
    -0.47
    POSITIVE LOGITS
     Bet
    0.36
     Secret
    0.35
     "\"
    0.35
     bet
    0.35
     اح
    0.34
    󠁴
    0.34
     đồ
    0.34
     desear
    0.33
     secret
    0.33
     Pis
    0.32
    Activations Density 0.030%

    No Known Activations