Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 52
    Prev
    Next
    INDEX
    Explanations

    references to a specific person named Lu

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

    mentions of the name "Lu"

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

    words starting with Lu

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    principalTable
    -0.68
    SequentialGroup
    -0.64
    """.
    -0.63
    AnchorStyles
    -0.62
    ."</
    -0.59
     Blech
    -0.57
    SOUNDBITE
    -0.56
    ãng
    -0.56
    pdev
    -0.55
     @}
    -0.55
    POSITIVE LOGITS
    Lu
    0.79
     Lu
    0.76
     doInBackground
    0.64
    ngdoc
    0.63
    OCCURRED
    0.60
    errHandler
    0.55
     lu
    0.53
    krivelse
    0.52
    ChromeDriver
    0.51
    XmlAccessType
    0.51
    Activations Density 0.012%

    No Known Activations