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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 14721
    Prev
    Next
    INDEX
    Explanations

    terms related to various forms of nobility, particularly focusing on "duke" and associated titles

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

    following "Du" or "du"

    np_acts-logits-general · gemini-2.0-flash

    words beginning 'du'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    FormTagHelper
    -0.66
     Params
    -0.61
    PullParser
    -0.60
     跳转至
    -0.58
     OCS
    -0.57
    Dent
    -0.57
     mdi
    -0.57
    PMailer
    -0.57
    */}
    -0.56
     SMI
    -0.55
    POSITIVE LOGITS
    du
    0.90
     du
    0.89
     Du
    0.82
    Duff
    0.76
     DU
    0.70
     Dup
    0.70
     DUP
    0.67
    Du
    0.66
     Duval
    0.65
    DU
    0.65
    Activations Density 0.123%

    No Known Activations