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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 8315
    Prev
    Next
    INDEX
    Explanations

    terms related to sexual abuse and misconduct

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

    Sexual abuse/misconduct

    np_acts-logits-general · gemini-2.0-flash

    sexual misconduct

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    DoubleQuotes
    -0.92
     onAnimation
    -0.78
    tagHelperRunner
    -0.76
     שוליים
    -0.70
     ब्रेकडाउन
    -0.69
    новниш
    -0.67
    GEBURTSDATUM
    -0.66
     AssemblyCulture
    -0.66
     CreateTagHelper
    -0.66
     onCreateView
    -0.63
    POSITIVE LOGITS
     sexual
    1.44
     sexually
    1.29
     Sexual
    1.25
    Sexual
    1.20
     sex
    1.16
     sexuelle
    1.09
     seksual
    1.03
    sexual
    1.02
     rape
    1.00
     sexuales
    0.98
    Activations Density 0.216%

    No Known Activations