© 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-2-2B
    7. Residual Stream - 16k
    9. 19-GEMMASCOPE-RES-16K
    11. 6993
    Prev
    Next
    INDEX
    Explanations

    phrases related to harmful speech and online conduct

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 19-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_19/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.19.hook_resid_post
    Hook Layer
    19
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     parad
    -0.58
     发表于
    -0.50
    Plen
    -0.49
    ύπ
    -0.48
     indisponible
    -0.47
    kuper
    -0.47
    Skocz
    -0.47
    Palmarès
    -0.47
    liculas
    -0.47
     verzichten
    -0.46
    POSITIVE LOGITS
     insults
    1.22
     slander
    1.20
     harassment
    1.10
     insulting
    1.07
     bullying
    1.07
     attacks
    1.06
     insult
    1.02
     defamation
    1.00
     hateful
    0.94
     hate
    0.94
    Activations Density 0.521%

    No Known Activations