© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 10287
    Prev
    Next
    INDEX
    Explanations

    instances of courtly or formal address and hierarchical interactions among characters

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

    formal honorifics and deferential address associated with royalty or nobility (courtly titles and respectful forms of address).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    formal titles and terms of address in royal or aristocratic settings, especially "Majesty," "Lord," "Lady," and honorific phrases.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    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
     judiciaire
    -0.46
    tagHelperRunner
    -0.43
     regioni
    -0.37
    kedés
    -0.36
     tarea
    -0.36
    Américaine
    -0.36
     zog
    -0.36
    Trichloroethane
    -0.36
     carreira
    -0.35
     career
    -0.34
    POSITIVE LOGITS
    initState
    0.62
    LookAnd
    0.58
    detectChanges
    0.58
     resourceCulture
    0.57
     gentleman
    0.56
     Administrativna
    0.54
     gentlemen
    0.52
    Kindly
    0.52
    posedge
    0.51
     Gentlemen
    0.51
    Activations Density 0.192%

    No Known Activations