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

    Neuronpedia

    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. 20-GEMMASCOPE-RES-16K
    11. 8255
    Prev
    Next
    INDEX
    Explanations

    instances of deception, manipulation, and fraudulent behavior

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     uitges
    -0.45
     gesteld
    -0.44
    forte
    -0.43
     geïsole
    -0.42
     behov
    -0.42
     bepaald
    -0.41
    Soma
    -0.40
     bijzonder
    -0.39
     فريبيس
    -0.39
     wünsche
    -0.39
    POSITIVE LOGITS
     trick
    1.04
     tricks
    0.96
     Tricks
    0.87
     tricked
    0.86
     scam
    0.85
     Trick
    0.85
     trucos
    0.81
     sneaky
    0.80
    trick
    0.79
     deception
    0.79
    Activations Density 0.266%

    No Known Activations